{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "# import zipfile\n",
        "\n",
        "# # with zipfile.ZipFile(\"drive/MyDrive/test.zip\",\"r\") as z:\n",
        "# #     z.extractall(\".\")\n",
        "# with zipfile.ZipFile(r\"D:/Rana shahzad/dataset100.zip\", \"r\") as z:\n",
        "#     z.extractall(\".\")\n",
        "\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from keras.layers import Activation\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "# from keras.optimizers import Adam\n",
        "# from keras.layers.normalization import BatchNormalization\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import (\n",
        "    BatchNormalization, SeparableConv2D, MaxPooling2D, Activation, Flatten, Dropout, Dense\n",
        ")\n",
        "# initializing the path to the dataset\n",
        "# print(train_path)\n",
        "import warnings\n",
        "\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "from keras.preprocessing import image as imagess\n",
        "import os"
      ],
      "metadata": {
        "id": "3Xp44-Ey6TOk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dLHeg6ry7Ew1",
        "outputId": "6b616b2e-3f5e-4e7c-9e99-d32aeca66ec8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_images_path = '/content/drive/MyDrive/RGBthree'\n",
        "labels = os.listdir(train_images_path)\n",
        "print(labels)\n",
        "\n",
        "all_images = []\n",
        "all_labels = []\n",
        "plt.figure(figsize=[5, 5])\n",
        "flag = False\n",
        "datadir='/content/drive/MyDrive/RGBthree'\n",
        "for label in labels:\n",
        "    images_name = os.listdir(train_images_path + \"/\" + label)\n",
        "    for file in images_name:\n",
        "        img = imagess.load_img(train_images_path + \"/\" + label + \"/\" + file, grayscale=False, target_size=(224, 224, 3))\n",
        "        image_array = imagess.img_to_array(img)\n",
        "        # print(image_array.shape)\n",
        "        all_images.append(image_array)\n",
        "        # all_images = np.array(all_images)\n",
        "        # print(all_images.shape)\n",
        "        all_labels.append(label)\n",
        "    print(\"Appending : \", label)\n",
        "print(\"Appended\")\n",
        "\n",
        "print(len(all_images))\n",
        "print(len(all_labels))\n",
        "all_images = np.array(all_images)\n",
        "print(np.array(all_labels).shape)\n",
        "single_image = all_images[0]\n",
        "print(single_image.shape)\n",
        "#\n",
        "# used to determine number of output classes:\n",
        "num_classes = len(all_labels)\n",
        "print(num_classes)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "BkNDFpt56TxI",
        "outputId": "d5668ec3-9c06-4e52-8be4-7d4931f4bd5a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['1', '2', '3', '4', '5', '6', '7', '8', '9']\n",
            "Appending :  1\n",
            "Appending :  2\n",
            "Appending :  3\n",
            "Appending :  4\n",
            "Appending :  5\n",
            "Appending :  6\n",
            "Appending :  7\n",
            "Appending :  8\n",
            "Appending :  9\n",
            "Appended\n",
            "442\n",
            "442\n",
            "(442,)\n",
            "(224, 224, 3)\n",
            "442\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "from keras.models import Sequential, Input, Model\n",
        "from keras.layers import Dense, Dropout, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "\n",
        "from keras.layers.advanced_activations import LeakyReLU\n",
        "from keras.regularizers import l2"
      ],
      "metadata": {
        "id": "TESZ4gJs6Tu2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "eg_lambda = 0.001\n",
        "dropout = True\n",
        "\n",
        "# load = 1\n",
        "# time_steps = 1000\n",
        "# subsample = 50\n",
        "input_shape = (224, 224, 3)\n",
        "model = Sequential()\n",
        "model.add(Conv2D(64, (5, 5), input_shape=input_shape, activation='relu', padding='same'))\n",
        "model.add(Conv2D(64, (5, 5), activation='relu', padding='same'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(128, (5, 5), activation='relu', padding='same'))\n",
        "model.add(Conv2D(128, (5, 5), activation='relu', padding='same'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(256, (3, 3), activation='relu', padding='same'))\n",
        "model.add(Conv2D(256, (3, 3), activation='relu', padding='same'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(9))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "adam = Adam(lr=0.0001)\n",
        "#\n",
        "model.compile(optimizer=adam, loss=\"categorical_crossentropy\", metrics=['mse', 'accuracy', tf.keras.metrics.AUC(),\n",
        "                                                                        tf.keras.metrics.BinaryAccuracy(),\n",
        "                                                                        'binary_crossentropy',\n",
        "                                                                        tf.keras.metrics.Precision(),\n",
        "                                                                        tf.keras.metrics.FalseNegatives(),\n",
        "                                                                        tf.keras.metrics.FalsePositives(),\n",
        "                                                                        tf.keras.metrics.TrueNegatives(),\n",
        "                                                                        tf.keras.metrics.TruePositives()\n",
        "                                                                        ])\n",
        "#\n",
        "model.summary()\n",
        "# model.save('/content/drive/MyDrive/melware_model_final.h5')\n",
        "# compiling the structure of the model\n",
        "# model.compile(Adam(lr=0.0001), loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "\n",
        "# In[10]:"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZLHzQTWS6TsO",
        "outputId": "507cc86f-328e-4557-9ea0-19daa4fb7a07"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 224, 224, 64)      4864      \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 224, 224, 64)      102464    \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 224, 224, 64)     256       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 112, 112, 64)     0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 112, 112, 128)     204928    \n",
            "                                                                 \n",
            " conv2d_3 (Conv2D)           (None, 112, 112, 128)     409728    \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 112, 112, 128)    512       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 56, 56, 128)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 56, 56, 256)       295168    \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 56, 56, 256)       590080    \n",
            "                                                                 \n",
            " batch_normalization_2 (Batc  (None, 56, 56, 256)      1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 28, 28, 256)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 200704)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 128)               25690240  \n",
            "                                                                 \n",
            " batch_normalization_3 (Batc  (None, 128)              512       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 128)               0         \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 9)                 1161      \n",
            "                                                                 \n",
            " activation_1 (Activation)   (None, 9)                 0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 27,300,937\n",
            "Trainable params: 27,299,785\n",
            "Non-trainable params: 1,152\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# model.save('/content/drive/MyDrive/melware_model_final.h5')"
      ],
      "metadata": {
        "id": "jAMOdu1_AvyK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.callbacks import ModelCheckpoint\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from keras.utils import np_utils\n",
        "\n",
        "le = LabelEncoder()\n",
        "y = le.fit_transform(all_labels)\n",
        "classes = list(le.classes_)\n",
        "y = np_utils.to_categorical(y, num_classes=len(labels))\n",
        "print(\"y : \", y.shape)\n",
        "X, Y = np.array(all_images), np.array(y)\n",
        "print(\"X.shape :  \", X.shape)\n",
        "print(\"Y.shape :  \", Y.shape)\n",
        "# print(\"len X : \",len(X),\"len Y : \",len(Y))\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.1, random_state=0)\n",
        "print(\"X_train : \", len(X_train), \"y_train : \", len(y_train))\n",
        "print(\"X_test : \", len(X_test), \"y_test : \", len(y_test))\n",
        "# y_train = (np.arange(len(all_labels)) == y_train[:])\n",
        "# y_test = (np.arange(len(all_labels)) == y_test[:])\n",
        "from keras import backend as K\n",
        "\n",
        "path_model = '/content/drive/MyDrive/checkpoints' #model.save('/content/drive/MyDrive/melware_model_final.h5')\n",
        "K.set_value(model.optimizer.lr, 1e-3)  # set the learning rate\n",
        "\n",
        "checkpoint = ModelCheckpoint(path_model + \"melware_model_final.h5\", monitor='loss', verbose=1, save_best_only=True, #change model name\n",
        "                                 mode='auto', save_freq= int(y.size / 100))\n",
        "#model.compile(optimizer='adam',loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=['accuracy'])\n",
        "# training the model\n",
        "h = model.fit(x=X_train,\n",
        "              y=y_train,\n",
        "              epochs=15,\n",
        "              validation_data=(X_test, y_test),\n",
        "              verbose=1,\n",
        "              shuffle=True,\n",
        "              callbacks=[checkpoint])\n",
        "# model.save('/content/drive/MyDrive/melware_model_final2.h5')\n",
        "\n",
        "accuracy = h.history['accuracy']\n",
        "val_accuracy = h.history['val_accuracy']\n",
        "loss = h.history['loss']\n",
        "val_loss = h.history['val_loss']\n",
        "auc = h.history['auc']\n",
        "binary_accuracy = h.history['binary_accuracy']\n",
        "binary_crossentropy = h.history['binary_crossentropy']\n",
        "precision = h.history['precision']\n",
        "false_negatives = h.history['false_negatives']\n",
        "false_positives = h.history['false_positives']\n",
        "true_negatives = h.history['true_negatives']\n",
        "true_positives = h.history['true_positives']\n",
        "val_binary_accuracy = h.history['val_binary_accuracy']\n",
        "val_binary_crossentropy = h.history['val_binary_crossentropy']\n",
        "val_precision = h.history['val_precision']\n",
        "val_false_negatives = h.history['val_false_negatives']\n",
        "val_false_positives = h.history['val_false_positives']\n",
        "val_true_negatives = h.history['val_true_negatives']\n",
        "val_true_positives = h.history['val_true_positives']\n",
        "epochs_list = range(len(accuracy))\n",
        "\n",
        "plt.plot(epochs_list, accuracy, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs_list, val_accuracy, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "plt.savefig(\"G:\\My Drive\\Results\\accuracy.png\")\n",
        "plt.clf()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, loss, 'bo', label='Training loss')\n",
        "plt.plot(epochs_list, val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training loss')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\loss.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, auc, 'bo', label='auc')\n",
        "plt.title('auc')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\auc.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, binary_accuracy, 'bo', label='binary_accuracy')\n",
        "plt.title('binary_accuracy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\binary_accuracy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, binary_crossentropy, 'bo', label='binary_crossentropy')\n",
        "plt.title('binary_crossentropy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\binary_crossentropy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, precision, 'bo', label='precision')\n",
        "plt.title('precision')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\precision.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, false_negatives, 'bo', label='false_negatives')\n",
        "plt.title('false_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\false_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, false_positives, 'bo', label='false_positives')\n",
        "plt.title('false_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\false_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, true_negatives, 'bo', label='true_negatives')\n",
        "plt.title('true_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\true_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, true_positives, 'bo', label='true_positives')\n",
        "plt.title('true_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\true_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_binary_accuracy, 'bo', label='val_binary_accuracy')\n",
        "plt.title('val_binary_accuracy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_binary_accuracy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_binary_crossentropy, 'bo', label='val_binary_crossentropy')\n",
        "plt.title('val_binary_crossentropy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_binary_crossentropy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_precision, 'bo', label='val_precision')\n",
        "plt.title('val_precision')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_precision.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_false_negatives, 'bo', label='val_false_negatives')\n",
        "plt.title('val_false_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_false_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_false_positives, 'bo', label='val_false_positives')\n",
        "plt.title('val_false_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_false_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_true_negatives, 'bo', label='val_true_negatives')\n",
        "plt.title('val_true_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_true_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_true_positives, 'bo', label='val_true_positives')\n",
        "plt.title('val_true_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_true_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "#\n",
        "# from keras.utils.vis_utils import plot_model\n",
        "# plot_model(model, to_file='model_plot.png', show_shapes=True,rankdir=\"TR\",expand_nested=True, show_layer_names=True,show_dtype= True)\n",
        "#\n",
        "# from google.colab import drive\n",
        "# drive.mount('/content/drive')\n",
        "#\n",
        "# !pip install visualkeras\n",
        "#\n",
        "# import visualkeras\n",
        "# visualkeras.layered_view(model)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "D0oYJsSb6Tpy",
        "outputId": "57150db5-c214-4fe9-a534-f675695ac457"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "y :  (442, 9)\n",
            "X.shape :   (442, 224, 224, 3)\n",
            "Y.shape :   (442, 9)\n",
            "X_train :  397 y_train :  397\n",
            "X_test :  45 y_test :  45\n",
            "Epoch 1/15\n",
            "13/13 [==============================] - 786s 59s/step - loss: 1.7786 - mse: 0.0850 - accuracy: 0.3804 - auc: 0.7819 - binary_accuracy: 0.8914 - binary_crossentropy: 0.7813 - precision: 0.5882 - false_negatives: 367.0000 - false_positives: 21.0000 - true_negatives: 3155.0000 - true_positives: 30.0000 - val_loss: 3.7169 - val_mse: 0.1250 - val_accuracy: 0.1111 - val_auc: 0.5532 - val_binary_accuracy: 0.8543 - val_binary_crossentropy: 1.6323 - val_precision: 0.1111 - val_false_negatives: 43.0000 - val_false_positives: 16.0000 - val_true_negatives: 344.0000 - val_true_positives: 2.0000\n",
            "Epoch 2/15\n",
            "13/13 [==============================] - 757s 58s/step - loss: 1.6321 - mse: 0.0799 - accuracy: 0.4307 - auc: 0.8244 - binary_accuracy: 0.8981 - binary_crossentropy: 0.7739 - precision: 0.7089 - false_negatives: 341.0000 - false_positives: 23.0000 - true_negatives: 3153.0000 - true_positives: 56.0000 - val_loss: 3.0068 - val_mse: 0.1155 - val_accuracy: 0.2000 - val_auc: 0.5441 - val_binary_accuracy: 0.8691 - val_binary_crossentropy: 0.9508 - val_precision: 0.1000 - val_false_negatives: 44.0000 - val_false_positives: 9.0000 - val_true_negatives: 351.0000 - val_true_positives: 1.0000\n",
            "Epoch 3/15\n",
            "12/13 [==========================>...] - ETA: 59s - loss: 1.6108 - mse: 0.0767 - accuracy: 0.4245 - auc: 0.8228 - binary_accuracy: 0.9025 - binary_crossentropy: 0.7830 - precision: 0.7701 - false_negatives: 317.0000 - false_positives: 20.0000 - true_negatives: 3052.0000 - true_positives: 67.0000 \n",
            "Epoch 3: loss improved from inf to 1.62567, saving model to /content/drive/MyDrive/checkpointsmelware_model_final.h5\n",
            "13/13 [==============================] - 788s 61s/step - loss: 1.6257 - mse: 0.0773 - accuracy: 0.4207 - auc: 0.8180 - binary_accuracy: 0.9023 - binary_crossentropy: 0.7839 - precision: 0.7667 - false_negatives: 328.0000 - false_positives: 21.0000 - true_negatives: 3155.0000 - true_positives: 69.0000 - val_loss: 3.0636 - val_mse: 0.1211 - val_accuracy: 0.0667 - val_auc: 0.5307 - val_binary_accuracy: 0.8617 - val_binary_crossentropy: 0.9174 - val_precision: 0.1333 - val_false_negatives: 43.0000 - val_false_positives: 13.0000 - val_true_negatives: 347.0000 - val_true_positives: 2.0000\n",
            "Epoch 4/15\n",
            "13/13 [==============================] - 765s 58s/step - loss: 1.5120 - mse: 0.0735 - accuracy: 0.4710 - auc: 0.8455 - binary_accuracy: 0.9034 - binary_crossentropy: 0.7790 - precision: 0.7407 - false_negatives: 317.0000 - false_positives: 28.0000 - true_negatives: 3148.0000 - true_positives: 80.0000 - val_loss: 2.6828 - val_mse: 0.1121 - val_accuracy: 0.1556 - val_auc: 0.5183 - val_binary_accuracy: 0.8741 - val_binary_crossentropy: 0.9046 - val_precision: 0.2000 - val_false_negatives: 43.0000 - val_false_positives: 8.0000 - val_true_negatives: 352.0000 - val_true_positives: 2.0000\n",
            "Epoch 5/15\n",
            "13/13 [==============================] - 758s 58s/step - loss: 1.4880 - mse: 0.0726 - accuracy: 0.4786 - auc: 0.8503 - binary_accuracy: 0.9062 - binary_crossentropy: 0.7929 - precision: 0.7500 - false_negatives: 304.0000 - false_positives: 31.0000 - true_negatives: 3145.0000 - true_positives: 93.0000 - val_loss: 2.4872 - val_mse: 0.1050 - val_accuracy: 0.1778 - val_auc: 0.5343 - val_binary_accuracy: 0.8815 - val_binary_crossentropy: 0.8211 - val_precision: 0.2857 - val_false_negatives: 43.0000 - val_false_positives: 5.0000 - val_true_negatives: 355.0000 - val_true_positives: 2.0000\n",
            "Epoch 6/15\n",
            "12/13 [==========================>...] - ETA: 59s - loss: 1.3372 - mse: 0.0655 - accuracy: 0.5156 - auc: 0.8840 - binary_accuracy: 0.9146 - binary_crossentropy: 0.7673 - precision: 0.8156 - false_negatives: 269.0000 - false_positives: 26.0000 - true_negatives: 3046.0000 - true_positives: 115.0000 \n",
            "Epoch 6: loss improved from 1.62567 to 1.36479, saving model to /content/drive/MyDrive/checkpointsmelware_model_final.h5\n",
            "13/13 [==============================] - 768s 59s/step - loss: 1.3648 - mse: 0.0666 - accuracy: 0.5088 - auc: 0.8778 - binary_accuracy: 0.9135 - binary_crossentropy: 0.7712 - precision: 0.8056 - false_negatives: 281.0000 - false_positives: 28.0000 - true_negatives: 3148.0000 - true_positives: 116.0000 - val_loss: 2.6680 - val_mse: 0.1098 - val_accuracy: 0.1556 - val_auc: 0.5255 - val_binary_accuracy: 0.8765 - val_binary_crossentropy: 0.8861 - val_precision: 0.1429 - val_false_negatives: 44.0000 - val_false_positives: 6.0000 - val_true_negatives: 354.0000 - val_true_positives: 1.0000\n",
            "Epoch 7/15\n",
            "13/13 [==============================] - 765s 59s/step - loss: 1.3637 - mse: 0.0666 - accuracy: 0.5365 - auc: 0.8746 - binary_accuracy: 0.9144 - binary_crossentropy: 0.7758 - precision: 0.7935 - false_negatives: 274.0000 - false_positives: 32.0000 - true_negatives: 3144.0000 - true_positives: 123.0000 - val_loss: 2.6356 - val_mse: 0.1100 - val_accuracy: 0.1111 - val_auc: 0.5125 - val_binary_accuracy: 0.8790 - val_binary_crossentropy: 0.8605 - val_precision: 0.1667 - val_false_negatives: 44.0000 - val_false_positives: 5.0000 - val_true_negatives: 355.0000 - val_true_positives: 1.0000\n",
            "Epoch 8/15\n",
            "13/13 [==============================] - 760s 58s/step - loss: 1.2808 - mse: 0.0624 - accuracy: 0.5290 - auc: 0.8926 - binary_accuracy: 0.9233 - binary_crossentropy: 0.7558 - precision: 0.8436 - false_negatives: 246.0000 - false_positives: 28.0000 - true_negatives: 3148.0000 - true_positives: 151.0000 - val_loss: 2.8448 - val_mse: 0.1109 - val_accuracy: 0.1778 - val_auc: 0.5176 - val_binary_accuracy: 0.8765 - val_binary_crossentropy: 1.0937 - val_precision: 0.2222 - val_false_negatives: 43.0000 - val_false_positives: 7.0000 - val_true_negatives: 353.0000 - val_true_positives: 2.0000\n",
            "Epoch 9/15\n",
            "12/13 [==========================>...] - ETA: 59s - loss: 1.1808 - mse: 0.0575 - accuracy: 0.5729 - auc: 0.9099 - binary_accuracy: 0.9297 - binary_crossentropy: 0.7555 - precision: 0.9123 - false_negatives: 228.0000 - false_positives: 15.0000 - true_negatives: 3057.0000 - true_positives: 156.0000 \n",
            "Epoch 9: loss improved from 1.36479 to 1.16823, saving model to /content/drive/MyDrive/checkpointsmelware_model_final.h5\n",
            "13/13 [==============================] - 778s 60s/step - loss: 1.1682 - mse: 0.0569 - accuracy: 0.5819 - auc: 0.9121 - binary_accuracy: 0.9306 - binary_crossentropy: 0.7543 - precision: 0.9162 - false_negatives: 233.0000 - false_positives: 15.0000 - true_negatives: 3161.0000 - true_positives: 164.0000 - val_loss: 3.0153 - val_mse: 0.1162 - val_accuracy: 0.1111 - val_auc: 0.5060 - val_binary_accuracy: 0.8716 - val_binary_crossentropy: 1.3146 - val_precision: 0.1818 - val_false_negatives: 43.0000 - val_false_positives: 9.0000 - val_true_negatives: 351.0000 - val_true_positives: 2.0000\n",
            "Epoch 10/15\n",
            "13/13 [==============================] - 765s 58s/step - loss: 1.1663 - mse: 0.0566 - accuracy: 0.5945 - auc: 0.9108 - binary_accuracy: 0.9286 - binary_crossentropy: 0.7570 - precision: 0.8480 - false_negatives: 224.0000 - false_positives: 31.0000 - true_negatives: 3145.0000 - true_positives: 173.0000 - val_loss: 2.9612 - val_mse: 0.1175 - val_accuracy: 0.1556 - val_auc: 0.5106 - val_binary_accuracy: 0.8691 - val_binary_crossentropy: 1.3074 - val_precision: 0.1667 - val_false_negatives: 43.0000 - val_false_positives: 10.0000 - val_true_negatives: 350.0000 - val_true_positives: 2.0000\n",
            "Epoch 11/15\n",
            "13/13 [==============================] - 760s 58s/step - loss: 1.0605 - mse: 0.0521 - accuracy: 0.6373 - auc: 0.9304 - binary_accuracy: 0.9365 - binary_crossentropy: 0.7435 - precision: 0.9009 - false_negatives: 206.0000 - false_positives: 21.0000 - true_negatives: 3155.0000 - true_positives: 191.0000 - val_loss: 2.7348 - val_mse: 0.1110 - val_accuracy: 0.1556 - val_auc: 0.4849 - val_binary_accuracy: 0.8765 - val_binary_crossentropy: 0.9335 - val_precision: 0.1429 - val_false_negatives: 44.0000 - val_false_positives: 6.0000 - val_true_negatives: 354.0000 - val_true_positives: 1.0000\n",
            "Epoch 12/15\n",
            "12/13 [==========================>...] - ETA: 59s - loss: 0.9689 - mse: 0.0464 - accuracy: 0.6667 - auc: 0.9394 - binary_accuracy: 0.9447 - binary_crossentropy: 0.7474 - precision: 0.9214 - false_negatives: 173.0000 - false_positives: 18.0000 - true_negatives: 3054.0000 - true_positives: 211.0000 \n",
            "Epoch 12: loss improved from 1.16823 to 0.99169, saving model to /content/drive/MyDrive/checkpointsmelware_model_final.h5\n",
            "13/13 [==============================] - 768s 59s/step - loss: 0.9917 - mse: 0.0473 - accuracy: 0.6599 - auc: 0.9352 - binary_accuracy: 0.9440 - binary_crossentropy: 0.7500 - precision: 0.9227 - false_negatives: 182.0000 - false_positives: 18.0000 - true_negatives: 3158.0000 - true_positives: 215.0000 - val_loss: 2.8422 - val_mse: 0.1115 - val_accuracy: 0.1556 - val_auc: 0.5009 - val_binary_accuracy: 0.8765 - val_binary_crossentropy: 0.9179 - val_precision: 0.3077 - val_false_negatives: 41.0000 - val_false_positives: 9.0000 - val_true_negatives: 351.0000 - val_true_positives: 4.0000\n",
            "Epoch 13/15\n",
            "13/13 [==============================] - 759s 58s/step - loss: 0.9237 - mse: 0.0453 - accuracy: 0.6977 - auc: 0.9482 - binary_accuracy: 0.9446 - binary_crossentropy: 0.7324 - precision: 0.9307 - false_negatives: 182.0000 - false_positives: 16.0000 - true_negatives: 3160.0000 - true_positives: 215.0000 - val_loss: 2.5216 - val_mse: 0.1058 - val_accuracy: 0.1333 - val_auc: 0.4925 - val_binary_accuracy: 0.8864 - val_binary_crossentropy: 0.7969 - val_precision: 0.3333 - val_false_negatives: 44.0000 - val_false_positives: 2.0000 - val_true_negatives: 358.0000 - val_true_positives: 1.0000\n",
            "Epoch 14/15\n",
            "13/13 [==============================] - 760s 58s/step - loss: 0.8003 - mse: 0.0393 - accuracy: 0.7355 - auc: 0.9659 - binary_accuracy: 0.9496 - binary_crossentropy: 0.7152 - precision: 0.9357 - false_negatives: 164.0000 - false_positives: 16.0000 - true_negatives: 3160.0000 - true_positives: 233.0000 - val_loss: 2.7610 - val_mse: 0.1124 - val_accuracy: 0.1111 - val_auc: 0.4986 - val_binary_accuracy: 0.8716 - val_binary_crossentropy: 0.8476 - val_precision: 0.0000e+00 - val_false_negatives: 45.0000 - val_false_positives: 7.0000 - val_true_negatives: 353.0000 - val_true_positives: 0.0000e+00\n",
            "Epoch 15/15\n",
            "12/13 [==========================>...] - ETA: 59s - loss: 0.7657 - mse: 0.0383 - accuracy: 0.7500 - auc: 0.9672 - binary_accuracy: 0.9511 - binary_crossentropy: 0.7036 - precision: 0.9317 - false_negatives: 152.0000 - false_positives: 17.0000 - true_negatives: 3055.0000 - true_positives: 232.0000 \n",
            "Epoch 15: loss improved from 0.99169 to 0.76367, saving model to /content/drive/MyDrive/checkpointsmelware_model_final.h5\n",
            "13/13 [==============================] - 763s 59s/step - loss: 0.7637 - mse: 0.0381 - accuracy: 0.7531 - auc: 0.9676 - binary_accuracy: 0.9513 - binary_crossentropy: 0.7056 - precision: 0.9339 - false_negatives: 157.0000 - false_positives: 17.0000 - true_negatives: 3159.0000 - true_positives: 240.0000 - val_loss: 2.6496 - val_mse: 0.1092 - val_accuracy: 0.0667 - val_auc: 0.4901 - val_binary_accuracy: 0.8864 - val_binary_crossentropy: 0.8072 - val_precision: 0.3333 - val_false_negatives: 44.0000 - val_false_positives: 2.0000 - val_true_negatives: 358.0000 - val_true_positives: 1.0000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n",
            "No handles with labels found to put in legend.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f8907732450>"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD8CAYAAAB0IB+mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAN4ElEQVR4nO3cf6jdd33H8efLJl1YjXUkV5DcaDKWTkM3sLt0HcLsqBtp/0j+cEgCxSmlAbfKmEXocFSpfzmZAyGbRlacgq3VP+SCkfzhKgUxkls6S5NSuYuduVXoNXb9p6Rttvf+OKfe4+1Nz7f3fu896f08HxC43+/53HPefLh53nPPr1QVkqTN702THkCStDEMviQ1wuBLUiMMviQ1wuBLUiMMviQ1Ymzwk9yf5NkkT1zm8iT5QpL5JI8nuaH/MSVJa9XlHv5XgAOvcfmtwL7hv6PAv659LElS38YGv6oeAX71GksOAV+tgVPAW5O8va8BJUn92NLDdewCzo8cLwzP/WL5wiRHGfwVwDXXXPNH73rXu3q4eUlqx6OPPvrLqppazff2EfzOquo4cBxgZmam5ubmNvLmJekNL8l/r/Z7+3iVzjPA7pHj6eE5SdIVpI/gzwIfGr5a5ybg+ap61cM5kqTJGvuQTpIHgJuBnUkWgE8BWwGq6ovACeA2YB54AfjIeg0rSVq9scGvqiNjLi/gb3qbSJIa8fLLL7OwsMDFixdfddm2bduYnp5m69atvd3ehj5pK0lasrCwwPbt29mzZw9Jfn2+qrhw4QILCwvs3bu3t9vzoxUkaUIuXrzIjh07fiP2AEnYsWPHivf818LgS9IELY/9uPNrYfAlqREGX5IaYfAlaYIGL3Tsfn4tDL4kTci2bdu4cOHCq+L+yqt0tm3b1uvt+bJMSZqQ6elpFhYWWFxcfNVlr7wOv08GX5ImZOvWrb2+zn4cH9KRpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqRKfgJzmQ5Kkk80nuWeHydyR5OMljSR5Pclv/o0qS1mJs8JNcBRwDbgX2A0eS7F+27B+Ah6rqPcBh4F/6HlSStDZd7uHfCMxX1bmqegl4EDi0bE0Bbxl+fS3w8/5GlCT1oUvwdwHnR44XhudGfRq4PckCcAL42EpXlORokrkkc4uLi6sYV5K0Wn09aXsE+EpVTQO3AV9L8qrrrqrjVTVTVTNTU1M93bQkqYsuwX8G2D1yPD08N+oO4CGAqvohsA3Y2ceAkqR+dAn+aWBfkr1JrmbwpOzssjU/A24BSPJuBsH3MRtJuoKMDX5VXQLuAk4CTzJ4Nc6ZJPclOThcdjdwZ5IfAw8AH66qWq+hJUmv35Yui6rqBIMnY0fP3Tvy9Vngvf2OJknqk++0laRGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJakSn4Cc5kOSpJPNJ7rnMmg8mOZvkTJKv9zumJGmttoxbkOQq4Bjw58ACcDrJbFWdHVmzD/h74L1V9VySt63XwJKk1elyD/9GYL6qzlXVS8CDwKFla+4EjlXVcwBV9Wy/Y0qS1qpL8HcB50eOF4bnRl0HXJfkB0lOJTmw0hUlOZpkLsnc4uLi6iaWJK1KX0/abgH2ATcDR4AvJ3nr8kVVdbyqZqpqZmpqqqebliR10SX4zwC7R46nh+dGLQCzVfVyVf0U+AmDXwCSpCtEl+CfBvYl2ZvkauAwMLtszbcZ3LsnyU4GD/Gc63FOSdIajQ1+VV0C7gJOAk8CD1XVmST3JTk4XHYSuJDkLPAw8ImqurBeQ0uSXr9U1URueGZmpubm5iZy25L0RpXk0aqaWc33+k5bSWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWpEp+AnOZDkqSTzSe55jXUfSFJJZvobUZLUh7HBT3IVcAy4FdgPHEmyf4V124G/BX7U95CSpLXrcg//RmC+qs5V1UvAg8ChFdZ9BvgscLHH+SRJPekS/F3A+ZHjheG5X0tyA7C7qr7zWleU5GiSuSRzi4uLr3tYSdLqrflJ2yRvAj4P3D1ubVUdr6qZqpqZmppa601Lkl6HLsF/Btg9cjw9PPeK7cD1wPeTPA3cBMz6xK0kXVm6BP80sC/J3iRXA4eB2VcurKrnq2pnVe2pqj3AKeBgVc2ty8SSpFUZG/yqugTcBZwEngQeqqozSe5LcnC9B5Qk9WNLl0VVdQI4sezcvZdZe/Pax5Ik9c132kpSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDWiU/CTHEjyVJL5JPescPnHk5xN8niS7yV5Z/+jSpLWYmzwk1wFHANuBfYDR5LsX7bsMWCmqv4Q+Bbwj30PKklamy738G8E5qvqXFW9BDwIHBpdUFUPV9ULw8NTwHS/Y0qS1qpL8HcB50eOF4bnLucO4LsrXZDkaJK5JHOLi4vdp5QkrVmvT9omuR2YAT630uVVdbyqZqpqZmpqqs+bliSNsaXDmmeA3SPH08NzvyHJ+4FPAu+rqhf7GU+S1Jcu9/BPA/uS7E1yNXAYmB1dkOQ9wJeAg1X1bP9jSpLWamzwq+oScBdwEngSeKiqziS5L8nB4bLPAW8GvpnkP5PMXubqJEkT0uUhHarqBHBi2bl7R75+f89zSZJ65jttJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRnYKf5ECSp5LMJ7lnhct/K8k3hpf/KMmevgeVJK3N2OAnuQo4BtwK7AeOJNm/bNkdwHNV9XvAPwOf7XtQSdLadLmHfyMwX1Xnquol4EHg0LI1h4B/H379LeCWJOlvTEnSWm3psGYXcH7keAH448utqapLSZ4HdgC/HF2U5ChwdHj4YpInVjP0JrSTZXvVMPdiiXuxxL1Y8vur/cYuwe9NVR0HjgMkmauqmY28/SuVe7HEvVjiXixxL5YkmVvt93Z5SOcZYPfI8fTw3IprkmwBrgUurHYoSVL/ugT/NLAvyd4kVwOHgdlla2aBvxp+/ZfAf1RV9TemJGmtxj6kM3xM/i7gJHAVcH9VnUlyHzBXVbPAvwFfSzIP/IrBL4Vxjq9h7s3GvVjiXixxL5a4F0tWvRfxjrgktcF32kpSIwy+JDVi3YPvxzIs6bAXH09yNsnjSb6X5J2TmHMjjNuLkXUfSFJJNu1L8rrsRZIPDn82ziT5+kbPuFE6/B95R5KHkzw2/H9y2yTmXG9J7k/y7OXeq5SBLwz36fEkN3S64qpat38MnuT9L+B3gauBHwP7l635a+CLw68PA99Yz5km9a/jXvwZ8NvDrz/a8l4M120HHgFOATOTnnuCPxf7gMeA3xkev23Sc09wL44DHx1+vR94etJzr9Ne/ClwA/DEZS6/DfguEOAm4Eddrne97+H7sQxLxu5FVT1cVS8MD08xeM/DZtTl5wLgMww+l+niRg63wbrsxZ3Asap6DqCqnt3gGTdKl70o4C3Dr68Ffr6B822YqnqEwSseL+cQ8NUaOAW8Ncnbx13vegd/pY9l2HW5NVV1CXjlYxk2my57MeoOBr/BN6OxezH8E3V3VX1nIwebgC4/F9cB1yX5QZJTSQ5s2HQbq8tefBq4PckCcAL42MaMdsV5vT0BNvijFdRNktuBGeB9k55lEpK8Cfg88OEJj3Kl2MLgYZ2bGfzV90iSP6iq/5noVJNxBPhKVf1Tkj9h8P6f66vq/yY92BvBet/D92MZlnTZC5K8H/gkcLCqXtyg2TbauL3YDlwPfD/J0wweo5zdpE/cdvm5WABmq+rlqvop8BMGvwA2my57cQfwEEBV/RDYxuCD1VrTqSfLrXfw/ViGJWP3Isl7gC8xiP1mfZwWxuxFVT1fVTurak9V7WHwfMbBqlr1h0Zdwbr8H/k2g3v3JNnJ4CGecxs55Abpshc/A24BSPJuBsFf3NAprwyzwIeGr9a5CXi+qn4x7pvW9SGdWr+PZXjD6bgXnwPeDHxz+Lz1z6rq4MSGXicd96IJHffiJPAXSc4C/wt8oqo23V/BHffibuDLSf6OwRO4H96MdxCTPMDgl/zO4fMVnwK2AlTVFxk8f3EbMA+8AHyk0/Vuwr2SJK3Ad9pKUiMMviQ1wuBLUiMMviQ1wuBLUiMMviQ1wuBLUiP+H2qgkGgttLe4AAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD8CAYAAAB0IB+mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAN4ElEQVR4nO3cf6jdd33H8efLJl1YjXUkV5DcaDKWTkM3sLt0HcLsqBtp/0j+cEgCxSmlAbfKmEXocFSpfzmZAyGbRlacgq3VP+SCkfzhKgUxkls6S5NSuYuduVXoNXb9p6Rttvf+OKfe4+1Nz7f3fu896f08HxC43+/53HPefLh53nPPr1QVkqTN702THkCStDEMviQ1wuBLUiMMviQ1wuBLUiMMviQ1Ymzwk9yf5NkkT1zm8iT5QpL5JI8nuaH/MSVJa9XlHv5XgAOvcfmtwL7hv6PAv659LElS38YGv6oeAX71GksOAV+tgVPAW5O8va8BJUn92NLDdewCzo8cLwzP/WL5wiRHGfwVwDXXXPNH73rXu3q4eUlqx6OPPvrLqppazff2EfzOquo4cBxgZmam5ubmNvLmJekNL8l/r/Z7+3iVzjPA7pHj6eE5SdIVpI/gzwIfGr5a5ybg+ap61cM5kqTJGvuQTpIHgJuBnUkWgE8BWwGq6ovACeA2YB54AfjIeg0rSVq9scGvqiNjLi/gb3qbSJIa8fLLL7OwsMDFixdfddm2bduYnp5m69atvd3ehj5pK0lasrCwwPbt29mzZw9Jfn2+qrhw4QILCwvs3bu3t9vzoxUkaUIuXrzIjh07fiP2AEnYsWPHivf818LgS9IELY/9uPNrYfAlqREGX5IaYfAlaYIGL3Tsfn4tDL4kTci2bdu4cOHCq+L+yqt0tm3b1uvt+bJMSZqQ6elpFhYWWFxcfNVlr7wOv08GX5ImZOvWrb2+zn4cH9KRpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqhMGXpEYYfElqRKfgJzmQ5Kkk80nuWeHydyR5OMljSR5Pclv/o0qS1mJs8JNcBRwDbgX2A0eS7F+27B+Ah6rqPcBh4F/6HlSStDZd7uHfCMxX1bmqegl4EDi0bE0Bbxl+fS3w8/5GlCT1oUvwdwHnR44XhudGfRq4PckCcAL42EpXlORokrkkc4uLi6sYV5K0Wn09aXsE+EpVTQO3AV9L8qrrrqrjVTVTVTNTU1M93bQkqYsuwX8G2D1yPD08N+oO4CGAqvohsA3Y2ceAkqR+dAn+aWBfkr1JrmbwpOzssjU/A24BSPJuBsH3MRtJuoKMDX5VXQLuAk4CTzJ4Nc6ZJPclOThcdjdwZ5IfAw8AH66qWq+hJUmv35Yui6rqBIMnY0fP3Tvy9Vngvf2OJknqk++0laRGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJaoTBl6RGGHxJakSn4Cc5kOSpJPNJ7rnMmg8mOZvkTJKv9zumJGmttoxbkOQq4Bjw58ACcDrJbFWdHVmzD/h74L1V9VySt63XwJKk1elyD/9GYL6qzlXVS8CDwKFla+4EjlXVcwBV9Wy/Y0qS1qpL8HcB50eOF4bnRl0HXJfkB0lOJTmw0hUlOZpkLsnc4uLi6iaWJK1KX0/abgH2ATcDR4AvJ3nr8kVVdbyqZqpqZmpqqqebliR10SX4zwC7R46nh+dGLQCzVfVyVf0U+AmDXwCSpCtEl+CfBvYl2ZvkauAwMLtszbcZ3LsnyU4GD/Gc63FOSdIajQ1+VV0C7gJOAk8CD1XVmST3JTk4XHYSuJDkLPAw8ImqurBeQ0uSXr9U1URueGZmpubm5iZy25L0RpXk0aqaWc33+k5bSWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWqEwZekRhh8SWpEp+AnOZDkqSTzSe55jXUfSFJJZvobUZLUh7HBT3IVcAy4FdgPHEmyf4V124G/BX7U95CSpLXrcg//RmC+qs5V1UvAg8ChFdZ9BvgscLHH+SRJPekS/F3A+ZHjheG5X0tyA7C7qr7zWleU5GiSuSRzi4uLr3tYSdLqrflJ2yRvAj4P3D1ubVUdr6qZqpqZmppa601Lkl6HLsF/Btg9cjw9PPeK7cD1wPeTPA3cBMz6xK0kXVm6BP80sC/J3iRXA4eB2VcurKrnq2pnVe2pqj3AKeBgVc2ty8SSpFUZG/yqugTcBZwEngQeqqozSe5LcnC9B5Qk9WNLl0VVdQI4sezcvZdZe/Pax5Ik9c132kpSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDXC4EtSIwy+JDWiU/CTHEjyVJL5JPescPnHk5xN8niS7yV5Z/+jSpLWYmzwk1wFHANuBfYDR5LsX7bsMWCmqv4Q+Bbwj30PKklamy738G8E5qvqXFW9BDwIHBpdUFUPV9ULw8NTwHS/Y0qS1qpL8HcB50eOF4bnLucO4LsrXZDkaJK5JHOLi4vdp5QkrVmvT9omuR2YAT630uVVdbyqZqpqZmpqqs+bliSNsaXDmmeA3SPH08NzvyHJ+4FPAu+rqhf7GU+S1Jcu9/BPA/uS7E1yNXAYmB1dkOQ9wJeAg1X1bP9jSpLWamzwq+oScBdwEngSeKiqziS5L8nB4bLPAW8GvpnkP5PMXubqJEkT0uUhHarqBHBi2bl7R75+f89zSZJ65jttJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRBl+SGmHwJakRnYKf5ECSp5LMJ7lnhct/K8k3hpf/KMmevgeVJK3N2OAnuQo4BtwK7AeOJNm/bNkdwHNV9XvAPwOf7XtQSdLadLmHfyMwX1Xnquol4EHg0LI1h4B/H379LeCWJOlvTEnSWm3psGYXcH7keAH448utqapLSZ4HdgC/HF2U5ChwdHj4YpInVjP0JrSTZXvVMPdiiXuxxL1Y8vur/cYuwe9NVR0HjgMkmauqmY28/SuVe7HEvVjiXixxL5YkmVvt93Z5SOcZYPfI8fTw3IprkmwBrgUurHYoSVL/ugT/NLAvyd4kVwOHgdlla2aBvxp+/ZfAf1RV9TemJGmtxj6kM3xM/i7gJHAVcH9VnUlyHzBXVbPAvwFfSzIP/IrBL4Vxjq9h7s3GvVjiXixxL5a4F0tWvRfxjrgktcF32kpSIwy+JDVi3YPvxzIs6bAXH09yNsnjSb6X5J2TmHMjjNuLkXUfSFJJNu1L8rrsRZIPDn82ziT5+kbPuFE6/B95R5KHkzw2/H9y2yTmXG9J7k/y7OXeq5SBLwz36fEkN3S64qpat38MnuT9L+B3gauBHwP7l635a+CLw68PA99Yz5km9a/jXvwZ8NvDrz/a8l4M120HHgFOATOTnnuCPxf7gMeA3xkev23Sc09wL44DHx1+vR94etJzr9Ne/ClwA/DEZS6/DfguEOAm4Eddrne97+H7sQxLxu5FVT1cVS8MD08xeM/DZtTl5wLgMww+l+niRg63wbrsxZ3Asap6DqCqnt3gGTdKl70o4C3Dr68Ffr6B822YqnqEwSseL+cQ8NUaOAW8Ncnbx13vegd/pY9l2HW5NVV1CXjlYxk2my57MeoOBr/BN6OxezH8E3V3VX1nIwebgC4/F9cB1yX5QZJTSQ5s2HQbq8tefBq4PckCcAL42MaMdsV5vT0BNvijFdRNktuBGeB9k55lEpK8Cfg88OEJj3Kl2MLgYZ2bGfzV90iSP6iq/5noVJNxBPhKVf1Tkj9h8P6f66vq/yY92BvBet/D92MZlnTZC5K8H/gkcLCqXtyg2TbauL3YDlwPfD/J0wweo5zdpE/cdvm5WABmq+rlqvop8BMGvwA2my57cQfwEEBV/RDYxuCD1VrTqSfLrXfw/ViGJWP3Isl7gC8xiP1mfZwWxuxFVT1fVTurak9V7WHwfMbBqlr1h0Zdwbr8H/k2g3v3JNnJ4CGecxs55Abpshc/A24BSPJuBsFf3NAprwyzwIeGr9a5CXi+qn4x7pvW9SGdWr+PZXjD6bgXnwPeDHxz+Lz1z6rq4MSGXicd96IJHffiJPAXSc4C/wt8oqo23V/BHffibuDLSf6OwRO4H96MdxCTPMDgl/zO4fMVnwK2AlTVFxk8f3EbMA+8AHyk0/Vuwr2SJK3Ad9pKUiMMviQ1wuBLUiMMviQ1wuBLUiMMviQ1wuBLUiP+H2qgkGgttLe4AAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "h.history\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5zusr2HYbeLK",
        "outputId": "7871b7f5-0a5a-4175-f267-8e9cc7812d94"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': [0.380352646112442,\n",
              "  0.4307304918766022,\n",
              "  0.42065492272377014,\n",
              "  0.4710327386856079,\n",
              "  0.47858941555023193,\n",
              "  0.508816123008728,\n",
              "  0.5365239381790161,\n",
              "  0.5289672613143921,\n",
              "  0.5818639993667603,\n",
              "  0.5944584608078003,\n",
              "  0.6372795701026917,\n",
              "  0.6599496006965637,\n",
              "  0.6977329850196838,\n",
              "  0.735516369342804,\n",
              "  0.75314861536026],\n",
              " 'auc': [0.7818894386291504,\n",
              "  0.824415922164917,\n",
              "  0.8180283308029175,\n",
              "  0.8454554080963135,\n",
              "  0.850271463394165,\n",
              "  0.8777846097946167,\n",
              "  0.8745847940444946,\n",
              "  0.8926350474357605,\n",
              "  0.9121230840682983,\n",
              "  0.9107728004455566,\n",
              "  0.9303632974624634,\n",
              "  0.9351515769958496,\n",
              "  0.9482476711273193,\n",
              "  0.9658601880073547,\n",
              "  0.9675878882408142],\n",
              " 'binary_accuracy': [0.89140784740448,\n",
              "  0.8981248140335083,\n",
              "  0.9023228883743286,\n",
              "  0.9034424424171448,\n",
              "  0.906241238117218,\n",
              "  0.9135180115699768,\n",
              "  0.9143576622009277,\n",
              "  0.9233137369155884,\n",
              "  0.9305905699729919,\n",
              "  0.9286313056945801,\n",
              "  0.9364678859710693,\n",
              "  0.9440246820449829,\n",
              "  0.9445844888687134,\n",
              "  0.9496220946311951,\n",
              "  0.9513013958930969],\n",
              " 'binary_crossentropy': [0.7812935709953308,\n",
              "  0.7738842964172363,\n",
              "  0.7839058041572571,\n",
              "  0.778957724571228,\n",
              "  0.7928581237792969,\n",
              "  0.7712150812149048,\n",
              "  0.7757919430732727,\n",
              "  0.7558392882347107,\n",
              "  0.7542902827262878,\n",
              "  0.7569898366928101,\n",
              "  0.7434684634208679,\n",
              "  0.7500337958335876,\n",
              "  0.7323569059371948,\n",
              "  0.7152118682861328,\n",
              "  0.7056466341018677],\n",
              " 'false_negatives': [367.0,\n",
              "  341.0,\n",
              "  328.0,\n",
              "  317.0,\n",
              "  304.0,\n",
              "  281.0,\n",
              "  274.0,\n",
              "  246.0,\n",
              "  233.0,\n",
              "  224.0,\n",
              "  206.0,\n",
              "  182.0,\n",
              "  182.0,\n",
              "  164.0,\n",
              "  157.0],\n",
              " 'false_positives': [21.0,\n",
              "  23.0,\n",
              "  21.0,\n",
              "  28.0,\n",
              "  31.0,\n",
              "  28.0,\n",
              "  32.0,\n",
              "  28.0,\n",
              "  15.0,\n",
              "  31.0,\n",
              "  21.0,\n",
              "  18.0,\n",
              "  16.0,\n",
              "  16.0,\n",
              "  17.0],\n",
              " 'loss': [1.7786270380020142,\n",
              "  1.6321122646331787,\n",
              "  1.6256695985794067,\n",
              "  1.5120073556900024,\n",
              "  1.4879672527313232,\n",
              "  1.3647894859313965,\n",
              "  1.3636817932128906,\n",
              "  1.280816912651062,\n",
              "  1.1682299375534058,\n",
              "  1.1663028001785278,\n",
              "  1.0605169534683228,\n",
              "  0.9916867613792419,\n",
              "  0.9237416386604309,\n",
              "  0.8003392219543457,\n",
              "  0.7636670470237732],\n",
              " 'mse': [0.08495943993330002,\n",
              "  0.07987001538276672,\n",
              "  0.07730439305305481,\n",
              "  0.07346060127019882,\n",
              "  0.07260838896036148,\n",
              "  0.06657734513282776,\n",
              "  0.06656047701835632,\n",
              "  0.062392544001340866,\n",
              "  0.056902043521404266,\n",
              "  0.05660899728536606,\n",
              "  0.052122484892606735,\n",
              "  0.04725772514939308,\n",
              "  0.04525408521294594,\n",
              "  0.03933892026543617,\n",
              "  0.03809608891606331],\n",
              " 'precision': [0.5882353186607361,\n",
              "  0.7088607549667358,\n",
              "  0.7666666507720947,\n",
              "  0.7407407164573669,\n",
              "  0.75,\n",
              "  0.8055555820465088,\n",
              "  0.7935484051704407,\n",
              "  0.8435754179954529,\n",
              "  0.916201114654541,\n",
              "  0.8480392098426819,\n",
              "  0.900943398475647,\n",
              "  0.9227467775344849,\n",
              "  0.9307359457015991,\n",
              "  0.935742974281311,\n",
              "  0.9338521361351013],\n",
              " 'true_negatives': [3155.0,\n",
              "  3153.0,\n",
              "  3155.0,\n",
              "  3148.0,\n",
              "  3145.0,\n",
              "  3148.0,\n",
              "  3144.0,\n",
              "  3148.0,\n",
              "  3161.0,\n",
              "  3145.0,\n",
              "  3155.0,\n",
              "  3158.0,\n",
              "  3160.0,\n",
              "  3160.0,\n",
              "  3159.0],\n",
              " 'true_positives': [30.0,\n",
              "  56.0,\n",
              "  69.0,\n",
              "  80.0,\n",
              "  93.0,\n",
              "  116.0,\n",
              "  123.0,\n",
              "  151.0,\n",
              "  164.0,\n",
              "  173.0,\n",
              "  191.0,\n",
              "  215.0,\n",
              "  215.0,\n",
              "  233.0,\n",
              "  240.0],\n",
              " 'val_accuracy': [0.1111111119389534,\n",
              "  0.20000000298023224,\n",
              "  0.06666667014360428,\n",
              "  0.15555556118488312,\n",
              "  0.17777778208255768,\n",
              "  0.15555556118488312,\n",
              "  0.1111111119389534,\n",
              "  0.17777778208255768,\n",
              "  0.1111111119389534,\n",
              "  0.15555556118488312,\n",
              "  0.15555556118488312,\n",
              "  0.15555556118488312,\n",
              "  0.13333334028720856,\n",
              "  0.1111111119389534,\n",
              "  0.06666667014360428],\n",
              " 'val_auc': [0.5531790256500244,\n",
              "  0.5441049933433533,\n",
              "  0.5307407975196838,\n",
              "  0.5183025002479553,\n",
              "  0.5343209505081177,\n",
              "  0.5255246758460999,\n",
              "  0.5124691128730774,\n",
              "  0.5176234841346741,\n",
              "  0.5060185194015503,\n",
              "  0.5105555653572083,\n",
              "  0.4848765432834625,\n",
              "  0.5008642673492432,\n",
              "  0.492530882358551,\n",
              "  0.49858027696609497,\n",
              "  0.49009260535240173],\n",
              " 'val_binary_accuracy': [0.8543209433555603,\n",
              "  0.869135856628418,\n",
              "  0.8617284297943115,\n",
              "  0.8740741014480591,\n",
              "  0.881481409072876,\n",
              "  0.8765431642532349,\n",
              "  0.879012405872345,\n",
              "  0.8765431642532349,\n",
              "  0.8716049194335938,\n",
              "  0.869135856628418,\n",
              "  0.8765431642532349,\n",
              "  0.8765431642532349,\n",
              "  0.8864196538925171,\n",
              "  0.8716049194335938,\n",
              "  0.8864196538925171],\n",
              " 'val_binary_crossentropy': [1.632258653640747,\n",
              "  0.9508339762687683,\n",
              "  0.917425274848938,\n",
              "  0.9046012163162231,\n",
              "  0.8211361169815063,\n",
              "  0.8860521912574768,\n",
              "  0.8604841232299805,\n",
              "  1.0937093496322632,\n",
              "  1.3145771026611328,\n",
              "  1.307363748550415,\n",
              "  0.933536946773529,\n",
              "  0.9179101586341858,\n",
              "  0.7969083189964294,\n",
              "  0.847583532333374,\n",
              "  0.8072296380996704],\n",
              " 'val_false_negatives': [43.0,\n",
              "  44.0,\n",
              "  43.0,\n",
              "  43.0,\n",
              "  43.0,\n",
              "  44.0,\n",
              "  44.0,\n",
              "  43.0,\n",
              "  43.0,\n",
              "  43.0,\n",
              "  44.0,\n",
              "  41.0,\n",
              "  44.0,\n",
              "  45.0,\n",
              "  44.0],\n",
              " 'val_false_positives': [16.0,\n",
              "  9.0,\n",
              "  13.0,\n",
              "  8.0,\n",
              "  5.0,\n",
              "  6.0,\n",
              "  5.0,\n",
              "  7.0,\n",
              "  9.0,\n",
              "  10.0,\n",
              "  6.0,\n",
              "  9.0,\n",
              "  2.0,\n",
              "  7.0,\n",
              "  2.0],\n",
              " 'val_loss': [3.716872215270996,\n",
              "  3.006753921508789,\n",
              "  3.0635550022125244,\n",
              "  2.6828250885009766,\n",
              "  2.4871675968170166,\n",
              "  2.667985439300537,\n",
              "  2.6355950832366943,\n",
              "  2.84482479095459,\n",
              "  3.0153019428253174,\n",
              "  2.961226463317871,\n",
              "  2.7347960472106934,\n",
              "  2.8421742916107178,\n",
              "  2.521578311920166,\n",
              "  2.7609620094299316,\n",
              "  2.649620771408081],\n",
              " 'val_mse': [0.1249580830335617,\n",
              "  0.11552278697490692,\n",
              "  0.12106549739837646,\n",
              "  0.11211096495389938,\n",
              "  0.10504983365535736,\n",
              "  0.10976365208625793,\n",
              "  0.1100398451089859,\n",
              "  0.11092673242092133,\n",
              "  0.11616352945566177,\n",
              "  0.11749245226383209,\n",
              "  0.11096540838479996,\n",
              "  0.11149732768535614,\n",
              "  0.1058126837015152,\n",
              "  0.1123923733830452,\n",
              "  0.10922669619321823],\n",
              " 'val_precision': [0.1111111119389534,\n",
              "  0.10000000149011612,\n",
              "  0.13333334028720856,\n",
              "  0.20000000298023224,\n",
              "  0.2857142984867096,\n",
              "  0.1428571492433548,\n",
              "  0.1666666716337204,\n",
              "  0.2222222238779068,\n",
              "  0.1818181872367859,\n",
              "  0.1666666716337204,\n",
              "  0.1428571492433548,\n",
              "  0.3076923191547394,\n",
              "  0.3333333432674408,\n",
              "  0.0,\n",
              "  0.3333333432674408],\n",
              " 'val_true_negatives': [344.0,\n",
              "  351.0,\n",
              "  347.0,\n",
              "  352.0,\n",
              "  355.0,\n",
              "  354.0,\n",
              "  355.0,\n",
              "  353.0,\n",
              "  351.0,\n",
              "  350.0,\n",
              "  354.0,\n",
              "  351.0,\n",
              "  358.0,\n",
              "  353.0,\n",
              "  358.0],\n",
              " 'val_true_positives': [2.0,\n",
              "  1.0,\n",
              "  2.0,\n",
              "  2.0,\n",
              "  2.0,\n",
              "  1.0,\n",
              "  1.0,\n",
              "  2.0,\n",
              "  2.0,\n",
              "  2.0,\n",
              "  1.0,\n",
              "  4.0,\n",
              "  1.0,\n",
              "  0.0,\n",
              "  1.0]}"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "accuracy = h.history['accuracy']\n",
        "val_accuracy = h.history['val_accuracy']\n",
        "loss = h.history['loss']\n",
        "val_loss = h.history['val_loss']\n",
        "auc = h.history['auc_2']\n",
        "binary_accuracy = h.history['binary_accuracy']\n",
        "binary_crossentropy = h.history['binary_crossentropy']\n",
        "precision = h.history['precision_2']\n",
        "false_negatives = h.history['false_negatives_2']\n",
        "false_positives = h.history['false_positives_2']\n",
        "true_negatives = h.history['true_negatives_2']\n",
        "true_positives = h.history['true_positives_2']\n",
        "val_binary_accuracy = h.history['val_binary_accuracy']\n",
        "val_binary_crossentropy = h.history['val_binary_crossentropy']\n",
        "val_precision = h.history['val_precision_2']\n",
        "val_false_negatives = h.history['val_false_negatives_2']\n",
        "val_false_positives = h.history['val_false_positives_2']\n",
        "val_true_negatives = h.history['val_true_negatives_2']\n",
        "val_true_positives = h.history['val_true_positives_2']\n",
        "epochs_list = range(len(accuracy))\n",
        "\n",
        "plt.plot(epochs_list, accuracy, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs_list, val_accuracy, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "plt.savefig(\"G:\\My Drive\\Results\\accuracy.png\")\n",
        "plt.clf()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, loss, 'bo', label='Training loss')\n",
        "plt.plot(epochs_list, val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training loss')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\loss.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, auc, 'bo', label='auc')\n",
        "plt.title('auc')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\auc.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, binary_accuracy, 'bo', label='binary_accuracy')\n",
        "plt.title('binary_accuracy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\binary_accuracy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, binary_crossentropy, 'bo', label='binary_crossentropy')\n",
        "plt.title('binary_crossentropy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\binary_crossentropy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, precision, 'bo', label='precision')\n",
        "plt.title('precision')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\precision.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, false_negatives, 'bo', label='false_negatives')\n",
        "plt.title('false_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\false_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, false_positives, 'bo', label='false_positives')\n",
        "plt.title('false_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\false_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, true_negatives, 'bo', label='true_negatives')\n",
        "plt.title('true_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\true_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, true_positives, 'bo', label='true_positives')\n",
        "plt.title('true_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\true_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_binary_accuracy, 'bo', label='val_binary_accuracy')\n",
        "plt.title('val_binary_accuracy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_binary_accuracy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_binary_crossentropy, 'bo', label='val_binary_crossentropy')\n",
        "plt.title('val_binary_crossentropy')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_binary_crossentropy.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_precision, 'bo', label='val_precision')\n",
        "plt.title('val_precision')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_precision.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_false_negatives, 'bo', label='val_false_negatives')\n",
        "plt.title('val_false_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_false_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_false_positives, 'bo', label='val_false_positives')\n",
        "plt.title('val_false_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_false_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_true_negatives, 'bo', label='val_true_negatives')\n",
        "plt.title('val_true_negatives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_true_negatives.png\")\n",
        "plt.clf()\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs_list, val_true_positives, 'bo', label='val_true_positives')\n",
        "plt.title('val_true_positives')\n",
        "plt.savefig(\"G:\\My Drive\\Results\\val_true_positives.png\")\n",
        "plt.clf()\n",
        "plt.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "id": "IcLFOxpYbfU9",
        "outputId": "ad5e6c25-1e72-4fd9-b988-88185d699198"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "KeyError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-11-547bb1fcb4c2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'loss'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mval_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'val_loss'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mauc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'auc_2'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0mbinary_accuracy\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'binary_accuracy'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0mbinary_crossentropy\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'binary_crossentropy'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyError\u001b[0m: 'auc_2'"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "Qp1qTmtZb6fC"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}