CSV_TRAINING_DATA.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.4"
    },
    "colab": {
      "name": "train-model.ipynb",
      "provenance": []
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "-AGLmQXOzU4b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "99af0776-0015-441c-ad99-2f2f00c1afc0"
      },
      "source": [
        "!git clone https://github.com/niazwazir/VTRG_Corona_Virus_Detection.git"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'VTRG_Corona_Virus_Detection'...\n",
            "remote: Enumerating objects: 222, done.\u001b[K\n",
            "remote: Counting objects: 100% (222/222), done.\u001b[K\n",
            "remote: Compressing objects: 100% (202/202), done.\u001b[K\n",
            "remote: Total 222 (delta 32), reused 174 (delta 14), pack-reused 0\u001b[K\n",
            "Receiving objects: 100% (222/222), 48.69 MiB | 12.70 MiB/s, done.\n",
            "Resolving deltas: 100% (32/32), done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i06B5KCXzpeA",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d8919913-f366-4be3-9bd5-b627ff3e8246"
      },
      "source": [
        "cd VTRG_Corona_Virus_Detection/"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/VTRG_Corona_Virus_Detection\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7RRNrOhfzQCA"
      },
      "source": [
        "#importing the necessary packages\n",
        "import numpy as np\n",
        "import cv2\n",
        "import os\n",
        "from imutils import paths\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.applications import VGG16, Xception, ResNet50\n",
        "from tensorflow.keras.layers import AveragePooling2D\n",
        "from tensorflow.keras.layers import Dropout\n",
        "from tensorflow.keras.layers import Flatten\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.layers import Input\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.preprocessing import LabelBinarizer\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.metrics import confusion_matrix"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TDMVsgutzQCC"
      },
      "source": [
        "#setting arguments \n",
        "args = {\"dataset\": \"./dataset\",\n",
        "        \"model\": \"corona.model\"\n",
        "       }"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U4aK9iI2zQCE"
      },
      "source": [
        "\"\"\"initializing the initial learning rate, epochs to train\n",
        "   and batch size\"\"\"\n",
        "learningRate = 1e-3\n",
        "epochs = 500\n",
        "batchSize = 8"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wMkv4ItzzQCG"
      },
      "source": [
        "\"\"\"listing out the images in the dataset folder, then initializing\n",
        "   the list of data (images) and class images\"\"\"\n",
        "imagePaths = list(paths.list_images(args[\"dataset\"]))\n",
        "data = []\n",
        "labels = []\n",
        "\n",
        "#looping over the images paths\n",
        "for imagePath in imagePaths:\n",
        "    \n",
        "    #extracting the class label from image paths\n",
        "    label = imagePath.split(os.path.sep)[-2]\n",
        "\n",
        "    \"\"\"loading the images, changing color channels and resizing it to be a fixed\n",
        "       224x224 pixels\"\"\"\n",
        "    image = cv2.imread(imagePath)\n",
        "    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
        "    image = cv2.resize(image, (224, 224))\n",
        "\n",
        "    #updating the data and labels lists\n",
        "    data.append(image)\n",
        "    labels.append(label)\n",
        "\n",
        "\"\"\"converting the data and labels to NumPy arrays and scaling the pixel\n",
        "   intensities to the range [0, 255]\"\"\"\n",
        "data = np.array(data) / 255.0\n",
        "labels = np.array(labels)"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "96ptpBK-zQCI"
      },
      "source": [
        "#using one-hot encoding on labels\n",
        "lb = LabelBinarizer()\n",
        "labels = lb.fit_transform(labels)\n",
        "labels = to_categorical(labels)\n",
        "\n",
        "\"\"\"splitting the data into training and testing using 80% of\n",
        "   the data for training and 20% for testing\"\"\"\n",
        "(trainX, testX, trainY, testY) = train_test_split(data, labels, test_size=0.20, stratify=labels, \n",
        "                                                  random_state=42)\n",
        "\n",
        "#initializing the training data augmentation object\n",
        "trainAug = ImageDataGenerator(\n",
        "    rotation_range=15,\n",
        "    fill_mode=\"nearest\")"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OADSafzSzQCK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "33c2af06-1b80-42c5-b5c7-97184a78e63e"
      },
      "source": [
        "#loading the VGG16 network, ensuring the head FC layer sets are left\n",
        "baseModel = VGG16(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(224, 224, 3)))\n",
        "#baseModel = Xception(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(224, 224, 3)))\n",
        "#baseModel = ResNet50(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(224, 224, 3)))\n",
        "\"\"\"constructing the main model that will be placed on top of the\n",
        "   the base model\"\"\"\n",
        "mainModel = baseModel.output\n",
        "\n",
        "mainModel = AveragePooling2D(pool_size=(4, 4))(mainModel)\n",
        "mainModel = Flatten(name=\"flatten\")(mainModel)\n",
        "mainModel = Dense(64, activation=\"relu\")(mainModel)\n",
        "mainModel = Dropout(0.5)(mainModel)\n",
        "mainModel = Dense(2, activation=\"softmax\")(mainModel)\n",
        "\n",
        "#lace the head FC model on top of the base model (actual model)\n",
        "model = Model(inputs=baseModel.input, outputs=mainModel)\n",
        "\n",
        "\"\"\"looping over all layers in the base model and freeze them so they will \n",
        "   not be updated during the first training process\"\"\"\n",
        "for layer in baseModel.layers:\n",
        "    layer.trainable = False"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "58892288/58889256 [==============================] - 2s 0us/step\n",
            "58900480/58889256 [==============================] - 2s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ubud0DSbrAgM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f362df43-24a8-4ea5-ee8d-a13824b3f240"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      \n",
            "                                                                 \n",
            " block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     \n",
            "                                                                 \n",
            " block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         \n",
            "                                                                 \n",
            " block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     \n",
            "                                                                 \n",
            " block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    \n",
            "                                                                 \n",
            " block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         \n",
            "                                                                 \n",
            " block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    \n",
            "                                                                 \n",
            " block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    \n",
            "                                                                 \n",
            " block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    \n",
            "                                                                 \n",
            " block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         \n",
            "                                                                 \n",
            " block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   \n",
            "                                                                 \n",
            " block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
            "                                                                 \n",
            " block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
            "                                                                 \n",
            " block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         \n",
            "                                                                 \n",
            " block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         \n",
            "                                                                 \n",
            " average_pooling2d (AverageP  (None, 1, 1, 512)        0         \n",
            " ooling2D)                                                       \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 512)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 64)                32832     \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 64)                0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 2)                 130       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 14,747,650\n",
            "Trainable params: 32,962\n",
            "Non-trainable params: 14,714,688\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "klehQ05Hz_4-"
      },
      "source": [
        "import keras\n",
        "import tensorflow as tf"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rp6QiHOoz_8L",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "ee67fb8b-0bdd-495e-c8b7-57c0b9c6d88a"
      },
      "source": [
        "tf.keras.utils.plot_model(model, 'Wazir_first_model.png')"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "filename='log.csv'\n",
        "history_logger=tf.keras.callbacks.CSVLogger(filename, separator=\",\", append=True)"
      ],
      "metadata": {
        "id": "O7pK3_ThbR7X"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#compiling up the model\n",
        "optimizer = Adam(lr=learningRate, decay=learningRate / epochs)\n",
        "model.compile(loss=\"binary_crossentropy\",optimizer=optimizer, metrics=[\"accuracy\"])"
      ],
      "metadata": {
        "id": "W71jr0kKdH-W",
        "outputId": "00078c3d-bd2e-412c-dabf-9992bfd2ff30",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "epochs = 500"
      ],
      "metadata": {
        "id": "nA_LqDiPdSHm"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history_const=model.fit(trainX, trainY,\n",
        "              batch_size=batchSize,\n",
        "              epochs=epochs,\n",
        "              callbacks=[history_logger],\n",
        "              validation_data=(trainX, trainY),\n",
        "              shuffle=True)"
      ],
      "metadata": {
        "id": "7NU3rDAIcHav",
        "outputId": "635afdbb-7f35-4ede-bfdf-9ce1d068b9b5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "12/12 [==============================] - 18s 392ms/step - loss: 0.7523 - accuracy: 0.5667 - val_loss: 0.6477 - val_accuracy: 0.6556\n",
            "Epoch 2/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.6165 - accuracy: 0.6444 - val_loss: 0.6189 - val_accuracy: 0.7889\n",
            "Epoch 3/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.6227 - accuracy: 0.6667 - val_loss: 0.5927 - val_accuracy: 0.7889\n",
            "Epoch 4/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.5612 - accuracy: 0.7000 - val_loss: 0.5675 - val_accuracy: 0.8111\n",
            "Epoch 5/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.5668 - accuracy: 0.7889 - val_loss: 0.5457 - val_accuracy: 0.7778\n",
            "Epoch 6/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.5328 - accuracy: 0.8111 - val_loss: 0.5194 - val_accuracy: 0.9333\n",
            "Epoch 7/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.5355 - accuracy: 0.7667 - val_loss: 0.4973 - val_accuracy: 0.8778\n",
            "Epoch 8/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.5244 - accuracy: 0.7778 - val_loss: 0.4707 - val_accuracy: 0.9333\n",
            "Epoch 9/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.4946 - accuracy: 0.8444 - val_loss: 0.4482 - val_accuracy: 0.9333\n",
            "Epoch 10/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.4716 - accuracy: 0.8556 - val_loss: 0.4286 - val_accuracy: 0.9556\n",
            "Epoch 11/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.4403 - accuracy: 0.8667 - val_loss: 0.4024 - val_accuracy: 0.9444\n",
            "Epoch 12/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.4063 - accuracy: 0.8667 - val_loss: 0.3885 - val_accuracy: 0.9333\n",
            "Epoch 13/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.4278 - accuracy: 0.8222 - val_loss: 0.3640 - val_accuracy: 0.9556\n",
            "Epoch 14/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.3947 - accuracy: 0.9222 - val_loss: 0.3495 - val_accuracy: 0.9556\n",
            "Epoch 15/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.3594 - accuracy: 0.9333 - val_loss: 0.3336 - val_accuracy: 0.9667\n",
            "Epoch 16/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.3646 - accuracy: 0.9222 - val_loss: 0.3145 - val_accuracy: 0.9444\n",
            "Epoch 17/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.3571 - accuracy: 0.9444 - val_loss: 0.3001 - val_accuracy: 0.9667\n",
            "Epoch 18/500\n",
            "12/12 [==============================] - 3s 242ms/step - loss: 0.3143 - accuracy: 0.9444 - val_loss: 0.2890 - val_accuracy: 0.9667\n",
            "Epoch 19/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.3221 - accuracy: 0.9222 - val_loss: 0.2757 - val_accuracy: 0.9556\n",
            "Epoch 20/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.2835 - accuracy: 0.9556 - val_loss: 0.2574 - val_accuracy: 0.9778\n",
            "Epoch 21/500\n",
            "12/12 [==============================] - 3s 246ms/step - loss: 0.2749 - accuracy: 0.9222 - val_loss: 0.2458 - val_accuracy: 0.9556\n",
            "Epoch 22/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.2615 - accuracy: 0.9333 - val_loss: 0.2360 - val_accuracy: 0.9667\n",
            "Epoch 23/500\n",
            "12/12 [==============================] - 3s 243ms/step - loss: 0.2619 - accuracy: 0.9444 - val_loss: 0.2274 - val_accuracy: 0.9778\n",
            "Epoch 24/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.2550 - accuracy: 0.9444 - val_loss: 0.2226 - val_accuracy: 0.9667\n",
            "Epoch 25/500\n",
            "12/12 [==============================] - 3s 252ms/step - loss: 0.2478 - accuracy: 0.9333 - val_loss: 0.2129 - val_accuracy: 0.9667\n",
            "Epoch 26/500\n",
            "12/12 [==============================] - 3s 244ms/step - loss: 0.2525 - accuracy: 0.9667 - val_loss: 0.2088 - val_accuracy: 0.9556\n",
            "Epoch 27/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.2372 - accuracy: 0.9333 - val_loss: 0.1990 - val_accuracy: 0.9778\n",
            "Epoch 28/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.2218 - accuracy: 0.9444 - val_loss: 0.1883 - val_accuracy: 0.9778\n",
            "Epoch 29/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.2079 - accuracy: 0.9667 - val_loss: 0.1829 - val_accuracy: 0.9778\n",
            "Epoch 30/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.2027 - accuracy: 0.9667 - val_loss: 0.1776 - val_accuracy: 0.9778\n",
            "Epoch 31/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1951 - accuracy: 0.9556 - val_loss: 0.1742 - val_accuracy: 0.9778\n",
            "Epoch 32/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.2166 - accuracy: 0.9444 - val_loss: 0.1692 - val_accuracy: 0.9778\n",
            "Epoch 33/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.2113 - accuracy: 0.9333 - val_loss: 0.1663 - val_accuracy: 0.9778\n",
            "Epoch 34/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.2097 - accuracy: 0.9667 - val_loss: 0.1605 - val_accuracy: 0.9778\n",
            "Epoch 35/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1928 - accuracy: 0.9556 - val_loss: 0.1588 - val_accuracy: 0.9778\n",
            "Epoch 36/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.2278 - accuracy: 0.9444 - val_loss: 0.1538 - val_accuracy: 0.9778\n",
            "Epoch 37/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1977 - accuracy: 0.9667 - val_loss: 0.1564 - val_accuracy: 0.9778\n",
            "Epoch 38/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1723 - accuracy: 0.9778 - val_loss: 0.1485 - val_accuracy: 0.9778\n",
            "Epoch 39/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1837 - accuracy: 0.9556 - val_loss: 0.1426 - val_accuracy: 0.9778\n",
            "Epoch 40/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1554 - accuracy: 0.9556 - val_loss: 0.1381 - val_accuracy: 0.9778\n",
            "Epoch 41/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1745 - accuracy: 0.9667 - val_loss: 0.1355 - val_accuracy: 0.9778\n",
            "Epoch 42/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1709 - accuracy: 0.9556 - val_loss: 0.1324 - val_accuracy: 0.9778\n",
            "Epoch 43/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1881 - accuracy: 0.9444 - val_loss: 0.1341 - val_accuracy: 0.9778\n",
            "Epoch 44/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1603 - accuracy: 0.9556 - val_loss: 0.1298 - val_accuracy: 0.9778\n",
            "Epoch 45/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1309 - accuracy: 0.9667 - val_loss: 0.1316 - val_accuracy: 0.9778\n",
            "Epoch 46/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1642 - accuracy: 0.9667 - val_loss: 0.1256 - val_accuracy: 0.9778\n",
            "Epoch 47/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1650 - accuracy: 0.9667 - val_loss: 0.1236 - val_accuracy: 0.9778\n",
            "Epoch 48/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1396 - accuracy: 0.9667 - val_loss: 0.1223 - val_accuracy: 0.9778\n",
            "Epoch 49/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1419 - accuracy: 0.9556 - val_loss: 0.1196 - val_accuracy: 0.9778\n",
            "Epoch 50/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1702 - accuracy: 0.9333 - val_loss: 0.1221 - val_accuracy: 0.9778\n",
            "Epoch 51/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1674 - accuracy: 0.9778 - val_loss: 0.1184 - val_accuracy: 0.9778\n",
            "Epoch 52/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1557 - accuracy: 0.9667 - val_loss: 0.1176 - val_accuracy: 0.9778\n",
            "Epoch 53/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1456 - accuracy: 0.9556 - val_loss: 0.1140 - val_accuracy: 0.9778\n",
            "Epoch 54/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1442 - accuracy: 0.9778 - val_loss: 0.1110 - val_accuracy: 0.9778\n",
            "Epoch 55/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1490 - accuracy: 0.9667 - val_loss: 0.1090 - val_accuracy: 0.9778\n",
            "Epoch 56/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1429 - accuracy: 0.9778 - val_loss: 0.1121 - val_accuracy: 0.9778\n",
            "Epoch 57/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1468 - accuracy: 0.9556 - val_loss: 0.1089 - val_accuracy: 0.9778\n",
            "Epoch 58/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.1628 - accuracy: 0.9556 - val_loss: 0.1058 - val_accuracy: 0.9778\n",
            "Epoch 59/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1483 - accuracy: 0.9667 - val_loss: 0.1089 - val_accuracy: 0.9778\n",
            "Epoch 60/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1361 - accuracy: 0.9778 - val_loss: 0.1077 - val_accuracy: 0.9778\n",
            "Epoch 61/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.1119 - accuracy: 0.9778 - val_loss: 0.1028 - val_accuracy: 0.9778\n",
            "Epoch 62/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1031 - accuracy: 0.9778 - val_loss: 0.0999 - val_accuracy: 0.9778\n",
            "Epoch 63/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1256 - accuracy: 0.9778 - val_loss: 0.0997 - val_accuracy: 0.9778\n",
            "Epoch 64/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0975 - accuracy: 0.9778 - val_loss: 0.0974 - val_accuracy: 0.9778\n",
            "Epoch 65/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.1288 - accuracy: 0.9556 - val_loss: 0.0974 - val_accuracy: 0.9778\n",
            "Epoch 66/500\n",
            "12/12 [==============================] - 3s 242ms/step - loss: 0.1234 - accuracy: 0.9667 - val_loss: 0.0989 - val_accuracy: 0.9778\n",
            "Epoch 67/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1004 - accuracy: 0.9778 - val_loss: 0.0955 - val_accuracy: 0.9778\n",
            "Epoch 68/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1017 - accuracy: 0.9778 - val_loss: 0.0943 - val_accuracy: 0.9778\n",
            "Epoch 69/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1312 - accuracy: 0.9667 - val_loss: 0.0953 - val_accuracy: 0.9778\n",
            "Epoch 70/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1157 - accuracy: 0.9778 - val_loss: 0.0927 - val_accuracy: 0.9778\n",
            "Epoch 71/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.1058 - accuracy: 0.9778 - val_loss: 0.0927 - val_accuracy: 0.9778\n",
            "Epoch 72/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.0977 - accuracy: 0.9556 - val_loss: 0.0903 - val_accuracy: 0.9778\n",
            "Epoch 73/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0932 - accuracy: 0.9778 - val_loss: 0.0903 - val_accuracy: 0.9778\n",
            "Epoch 74/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1252 - accuracy: 0.9778 - val_loss: 0.0890 - val_accuracy: 0.9778\n",
            "Epoch 75/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1091 - accuracy: 0.9778 - val_loss: 0.0880 - val_accuracy: 0.9778\n",
            "Epoch 76/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1062 - accuracy: 0.9667 - val_loss: 0.0867 - val_accuracy: 0.9778\n",
            "Epoch 77/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1248 - accuracy: 0.9667 - val_loss: 0.0870 - val_accuracy: 0.9778\n",
            "Epoch 78/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1294 - accuracy: 0.9667 - val_loss: 0.0875 - val_accuracy: 0.9778\n",
            "Epoch 79/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1165 - accuracy: 0.9556 - val_loss: 0.0858 - val_accuracy: 0.9778\n",
            "Epoch 80/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1155 - accuracy: 0.9778 - val_loss: 0.0863 - val_accuracy: 0.9778\n",
            "Epoch 81/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1234 - accuracy: 0.9667 - val_loss: 0.0843 - val_accuracy: 0.9778\n",
            "Epoch 82/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.0940 - accuracy: 0.9778 - val_loss: 0.0856 - val_accuracy: 0.9778\n",
            "Epoch 83/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1108 - accuracy: 0.9667 - val_loss: 0.0839 - val_accuracy: 0.9778\n",
            "Epoch 84/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.0900 - accuracy: 0.9778 - val_loss: 0.0826 - val_accuracy: 0.9778\n",
            "Epoch 85/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.1161 - accuracy: 0.9667 - val_loss: 0.0822 - val_accuracy: 0.9778\n",
            "Epoch 86/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1099 - accuracy: 0.9778 - val_loss: 0.0795 - val_accuracy: 0.9778\n",
            "Epoch 87/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1084 - accuracy: 0.9778 - val_loss: 0.0809 - val_accuracy: 0.9778\n",
            "Epoch 88/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0874 - accuracy: 0.9778 - val_loss: 0.0806 - val_accuracy: 0.9778\n",
            "Epoch 89/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0975 - accuracy: 0.9778 - val_loss: 0.0801 - val_accuracy: 0.9778\n",
            "Epoch 90/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0992 - accuracy: 0.9667 - val_loss: 0.0791 - val_accuracy: 0.9778\n",
            "Epoch 91/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1206 - accuracy: 0.9667 - val_loss: 0.0781 - val_accuracy: 0.9778\n",
            "Epoch 92/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0944 - accuracy: 0.9667 - val_loss: 0.0785 - val_accuracy: 0.9778\n",
            "Epoch 93/500\n",
            "12/12 [==============================] - 3s 233ms/step - loss: 0.0970 - accuracy: 0.9778 - val_loss: 0.0771 - val_accuracy: 0.9778\n",
            "Epoch 94/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0967 - accuracy: 0.9778 - val_loss: 0.0768 - val_accuracy: 0.9778\n",
            "Epoch 95/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0864 - accuracy: 0.9778 - val_loss: 0.0775 - val_accuracy: 0.9778\n",
            "Epoch 96/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1012 - accuracy: 0.9778 - val_loss: 0.0750 - val_accuracy: 0.9778\n",
            "Epoch 97/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1016 - accuracy: 0.9778 - val_loss: 0.0746 - val_accuracy: 0.9778\n",
            "Epoch 98/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0823 - accuracy: 0.9778 - val_loss: 0.0729 - val_accuracy: 0.9778\n",
            "Epoch 99/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0786 - accuracy: 0.9889 - val_loss: 0.0716 - val_accuracy: 0.9778\n",
            "Epoch 100/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0881 - accuracy: 0.9778 - val_loss: 0.0724 - val_accuracy: 0.9778\n",
            "Epoch 101/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.1074 - accuracy: 0.9778 - val_loss: 0.0717 - val_accuracy: 0.9778\n",
            "Epoch 102/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0988 - accuracy: 0.9778 - val_loss: 0.0701 - val_accuracy: 0.9778\n",
            "Epoch 103/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0815 - accuracy: 0.9778 - val_loss: 0.0705 - val_accuracy: 0.9778\n",
            "Epoch 104/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0917 - accuracy: 0.9667 - val_loss: 0.0718 - val_accuracy: 0.9778\n",
            "Epoch 105/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0920 - accuracy: 0.9778 - val_loss: 0.0717 - val_accuracy: 0.9778\n",
            "Epoch 106/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1043 - accuracy: 0.9778 - val_loss: 0.0703 - val_accuracy: 0.9778\n",
            "Epoch 107/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.1055 - accuracy: 0.9667 - val_loss: 0.0703 - val_accuracy: 0.9778\n",
            "Epoch 108/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0892 - accuracy: 0.9778 - val_loss: 0.0697 - val_accuracy: 0.9778\n",
            "Epoch 109/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.1027 - accuracy: 0.9778 - val_loss: 0.0699 - val_accuracy: 0.9778\n",
            "Epoch 110/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0939 - accuracy: 0.9778 - val_loss: 0.0680 - val_accuracy: 0.9778\n",
            "Epoch 111/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.1158 - accuracy: 0.9667 - val_loss: 0.0702 - val_accuracy: 0.9778\n",
            "Epoch 112/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.0930 - accuracy: 0.9778 - val_loss: 0.0676 - val_accuracy: 0.9778\n",
            "Epoch 113/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1027 - accuracy: 0.9667 - val_loss: 0.0662 - val_accuracy: 0.9778\n",
            "Epoch 114/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0845 - accuracy: 0.9778 - val_loss: 0.0679 - val_accuracy: 0.9778\n",
            "Epoch 115/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0909 - accuracy: 0.9778 - val_loss: 0.0664 - val_accuracy: 0.9778\n",
            "Epoch 116/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0748 - accuracy: 0.9778 - val_loss: 0.0645 - val_accuracy: 0.9778\n",
            "Epoch 117/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0776 - accuracy: 0.9889 - val_loss: 0.0634 - val_accuracy: 0.9778\n",
            "Epoch 118/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.0981 - accuracy: 0.9667 - val_loss: 0.0627 - val_accuracy: 0.9778\n",
            "Epoch 119/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0997 - accuracy: 0.9778 - val_loss: 0.0630 - val_accuracy: 0.9778\n",
            "Epoch 120/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0769 - accuracy: 0.9667 - val_loss: 0.0625 - val_accuracy: 0.9778\n",
            "Epoch 121/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1229 - accuracy: 0.9667 - val_loss: 0.0691 - val_accuracy: 0.9778\n",
            "Epoch 122/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0909 - accuracy: 0.9778 - val_loss: 0.0665 - val_accuracy: 0.9778\n",
            "Epoch 123/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0955 - accuracy: 0.9667 - val_loss: 0.0634 - val_accuracy: 0.9778\n",
            "Epoch 124/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0808 - accuracy: 0.9667 - val_loss: 0.0634 - val_accuracy: 0.9778\n",
            "Epoch 125/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0739 - accuracy: 0.9778 - val_loss: 0.0629 - val_accuracy: 0.9778\n",
            "Epoch 126/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0590 - accuracy: 0.9778 - val_loss: 0.0600 - val_accuracy: 0.9778\n",
            "Epoch 127/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0668 - accuracy: 0.9778 - val_loss: 0.0600 - val_accuracy: 0.9778\n",
            "Epoch 128/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0764 - accuracy: 0.9667 - val_loss: 0.0610 - val_accuracy: 0.9778\n",
            "Epoch 129/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0893 - accuracy: 0.9667 - val_loss: 0.0599 - val_accuracy: 0.9778\n",
            "Epoch 130/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0581 - accuracy: 0.9778 - val_loss: 0.0646 - val_accuracy: 0.9778\n",
            "Epoch 131/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1136 - accuracy: 0.9556 - val_loss: 0.0592 - val_accuracy: 0.9778\n",
            "Epoch 132/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0664 - accuracy: 0.9778 - val_loss: 0.0592 - val_accuracy: 0.9778\n",
            "Epoch 133/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0855 - accuracy: 0.9667 - val_loss: 0.0578 - val_accuracy: 0.9778\n",
            "Epoch 134/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0910 - accuracy: 0.9778 - val_loss: 0.0572 - val_accuracy: 0.9778\n",
            "Epoch 135/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.0788 - accuracy: 0.9667 - val_loss: 0.0573 - val_accuracy: 0.9778\n",
            "Epoch 136/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0803 - accuracy: 0.9778 - val_loss: 0.0573 - val_accuracy: 0.9778\n",
            "Epoch 137/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0884 - accuracy: 0.9778 - val_loss: 0.0565 - val_accuracy: 0.9778\n",
            "Epoch 138/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0701 - accuracy: 0.9778 - val_loss: 0.0564 - val_accuracy: 0.9778\n",
            "Epoch 139/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0820 - accuracy: 0.9889 - val_loss: 0.0566 - val_accuracy: 0.9778\n",
            "Epoch 140/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0867 - accuracy: 0.9778 - val_loss: 0.0574 - val_accuracy: 0.9778\n",
            "Epoch 141/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0696 - accuracy: 0.9778 - val_loss: 0.0590 - val_accuracy: 0.9778\n",
            "Epoch 142/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.1020 - accuracy: 0.9667 - val_loss: 0.0578 - val_accuracy: 0.9778\n",
            "Epoch 143/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0693 - accuracy: 0.9889 - val_loss: 0.0577 - val_accuracy: 0.9778\n",
            "Epoch 144/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0679 - accuracy: 0.9778 - val_loss: 0.0553 - val_accuracy: 0.9778\n",
            "Epoch 145/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0729 - accuracy: 0.9778 - val_loss: 0.0543 - val_accuracy: 0.9778\n",
            "Epoch 146/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0633 - accuracy: 0.9778 - val_loss: 0.0574 - val_accuracy: 0.9778\n",
            "Epoch 147/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0808 - accuracy: 0.9889 - val_loss: 0.0533 - val_accuracy: 0.9778\n",
            "Epoch 148/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0965 - accuracy: 0.9667 - val_loss: 0.0536 - val_accuracy: 0.9778\n",
            "Epoch 149/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0824 - accuracy: 0.9667 - val_loss: 0.0529 - val_accuracy: 0.9778\n",
            "Epoch 150/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0654 - accuracy: 0.9667 - val_loss: 0.0578 - val_accuracy: 0.9667\n",
            "Epoch 151/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0854 - accuracy: 0.9778 - val_loss: 0.0536 - val_accuracy: 0.9778\n",
            "Epoch 152/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0580 - accuracy: 0.9778 - val_loss: 0.0528 - val_accuracy: 0.9778\n",
            "Epoch 153/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0847 - accuracy: 0.9778 - val_loss: 0.0538 - val_accuracy: 0.9778\n",
            "Epoch 154/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0908 - accuracy: 0.9556 - val_loss: 0.0529 - val_accuracy: 0.9778\n",
            "Epoch 155/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.0674 - accuracy: 0.9778 - val_loss: 0.0531 - val_accuracy: 0.9778\n",
            "Epoch 156/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0739 - accuracy: 0.9667 - val_loss: 0.0519 - val_accuracy: 0.9778\n",
            "Epoch 157/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0783 - accuracy: 0.9556 - val_loss: 0.0547 - val_accuracy: 0.9778\n",
            "Epoch 158/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0616 - accuracy: 0.9778 - val_loss: 0.0531 - val_accuracy: 0.9778\n",
            "Epoch 159/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0580 - accuracy: 0.9778 - val_loss: 0.0530 - val_accuracy: 0.9778\n",
            "Epoch 160/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0786 - accuracy: 0.9778 - val_loss: 0.0531 - val_accuracy: 0.9778\n",
            "Epoch 161/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0766 - accuracy: 0.9778 - val_loss: 0.0516 - val_accuracy: 0.9778\n",
            "Epoch 162/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0549 - accuracy: 0.9778 - val_loss: 0.0504 - val_accuracy: 0.9778\n",
            "Epoch 163/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0702 - accuracy: 0.9667 - val_loss: 0.0495 - val_accuracy: 0.9778\n",
            "Epoch 164/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0674 - accuracy: 0.9778 - val_loss: 0.0500 - val_accuracy: 0.9778\n",
            "Epoch 165/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0649 - accuracy: 0.9778 - val_loss: 0.0491 - val_accuracy: 0.9778\n",
            "Epoch 166/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0725 - accuracy: 0.9778 - val_loss: 0.0513 - val_accuracy: 0.9778\n",
            "Epoch 167/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0602 - accuracy: 0.9667 - val_loss: 0.0487 - val_accuracy: 0.9778\n",
            "Epoch 168/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0704 - accuracy: 0.9778 - val_loss: 0.0487 - val_accuracy: 0.9778\n",
            "Epoch 169/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0613 - accuracy: 0.9778 - val_loss: 0.0482 - val_accuracy: 0.9778\n",
            "Epoch 170/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0688 - accuracy: 0.9778 - val_loss: 0.0485 - val_accuracy: 0.9778\n",
            "Epoch 171/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0484 - accuracy: 0.9889 - val_loss: 0.0483 - val_accuracy: 0.9778\n",
            "Epoch 172/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0539 - accuracy: 0.9778 - val_loss: 0.0487 - val_accuracy: 0.9778\n",
            "Epoch 173/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0796 - accuracy: 0.9778 - val_loss: 0.0474 - val_accuracy: 0.9778\n",
            "Epoch 174/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0611 - accuracy: 0.9778 - val_loss: 0.0517 - val_accuracy: 0.9778\n",
            "Epoch 175/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0672 - accuracy: 0.9667 - val_loss: 0.0476 - val_accuracy: 0.9778\n",
            "Epoch 176/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0560 - accuracy: 0.9889 - val_loss: 0.0473 - val_accuracy: 0.9778\n",
            "Epoch 177/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0641 - accuracy: 0.9778 - val_loss: 0.0470 - val_accuracy: 0.9778\n",
            "Epoch 178/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0613 - accuracy: 0.9778 - val_loss: 0.0471 - val_accuracy: 0.9778\n",
            "Epoch 179/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0662 - accuracy: 0.9667 - val_loss: 0.0473 - val_accuracy: 0.9778\n",
            "Epoch 180/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0614 - accuracy: 0.9778 - val_loss: 0.0487 - val_accuracy: 0.9778\n",
            "Epoch 181/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0601 - accuracy: 0.9778 - val_loss: 0.0475 - val_accuracy: 0.9778\n",
            "Epoch 182/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0594 - accuracy: 0.9889 - val_loss: 0.0466 - val_accuracy: 0.9778\n",
            "Epoch 183/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0634 - accuracy: 0.9778 - val_loss: 0.0467 - val_accuracy: 0.9778\n",
            "Epoch 184/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0542 - accuracy: 0.9778 - val_loss: 0.0517 - val_accuracy: 0.9667\n",
            "Epoch 185/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0629 - accuracy: 0.9667 - val_loss: 0.0485 - val_accuracy: 0.9778\n",
            "Epoch 186/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0604 - accuracy: 0.9778 - val_loss: 0.0479 - val_accuracy: 0.9778\n",
            "Epoch 187/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0597 - accuracy: 0.9667 - val_loss: 0.0463 - val_accuracy: 0.9778\n",
            "Epoch 188/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0629 - accuracy: 0.9778 - val_loss: 0.0464 - val_accuracy: 0.9778\n",
            "Epoch 189/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0488 - accuracy: 1.0000 - val_loss: 0.0465 - val_accuracy: 0.9778\n",
            "Epoch 190/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0476 - accuracy: 0.9889 - val_loss: 0.0462 - val_accuracy: 0.9778\n",
            "Epoch 191/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0697 - accuracy: 0.9667 - val_loss: 0.0467 - val_accuracy: 0.9778\n",
            "Epoch 192/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0541 - accuracy: 0.9778 - val_loss: 0.0498 - val_accuracy: 0.9778\n",
            "Epoch 193/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0837 - accuracy: 0.9778 - val_loss: 0.0468 - val_accuracy: 0.9778\n",
            "Epoch 194/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0539 - accuracy: 0.9778 - val_loss: 0.0455 - val_accuracy: 0.9778\n",
            "Epoch 195/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0720 - accuracy: 0.9778 - val_loss: 0.0457 - val_accuracy: 0.9778\n",
            "Epoch 196/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0561 - accuracy: 0.9778 - val_loss: 0.0458 - val_accuracy: 0.9778\n",
            "Epoch 197/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0675 - accuracy: 0.9667 - val_loss: 0.0453 - val_accuracy: 0.9778\n",
            "Epoch 198/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0557 - accuracy: 0.9778 - val_loss: 0.0459 - val_accuracy: 0.9778\n",
            "Epoch 199/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0613 - accuracy: 0.9778 - val_loss: 0.0444 - val_accuracy: 0.9778\n",
            "Epoch 200/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0672 - accuracy: 0.9667 - val_loss: 0.0443 - val_accuracy: 0.9778\n",
            "Epoch 201/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0600 - accuracy: 0.9667 - val_loss: 0.0441 - val_accuracy: 0.9778\n",
            "Epoch 202/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0482 - accuracy: 0.9778 - val_loss: 0.0447 - val_accuracy: 0.9778\n",
            "Epoch 203/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0505 - accuracy: 0.9778 - val_loss: 0.0464 - val_accuracy: 0.9778\n",
            "Epoch 204/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0726 - accuracy: 0.9778 - val_loss: 0.0450 - val_accuracy: 0.9778\n",
            "Epoch 205/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0598 - accuracy: 0.9667 - val_loss: 0.0454 - val_accuracy: 0.9778\n",
            "Epoch 206/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0540 - accuracy: 0.9778 - val_loss: 0.0445 - val_accuracy: 0.9778\n",
            "Epoch 207/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0666 - accuracy: 0.9667 - val_loss: 0.0443 - val_accuracy: 0.9778\n",
            "Epoch 208/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0657 - accuracy: 0.9667 - val_loss: 0.0431 - val_accuracy: 0.9778\n",
            "Epoch 209/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0735 - accuracy: 0.9667 - val_loss: 0.0436 - val_accuracy: 0.9778\n",
            "Epoch 210/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0696 - accuracy: 0.9778 - val_loss: 0.0463 - val_accuracy: 0.9778\n",
            "Epoch 211/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0423 - accuracy: 0.9778 - val_loss: 0.0457 - val_accuracy: 0.9778\n",
            "Epoch 212/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0542 - accuracy: 0.9667 - val_loss: 0.0435 - val_accuracy: 0.9778\n",
            "Epoch 213/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0532 - accuracy: 0.9889 - val_loss: 0.0449 - val_accuracy: 0.9778\n",
            "Epoch 214/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0605 - accuracy: 0.9889 - val_loss: 0.0443 - val_accuracy: 0.9778\n",
            "Epoch 215/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0913 - accuracy: 0.9556 - val_loss: 0.0446 - val_accuracy: 0.9778\n",
            "Epoch 216/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0633 - accuracy: 0.9556 - val_loss: 0.0447 - val_accuracy: 0.9778\n",
            "Epoch 217/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0657 - accuracy: 0.9889 - val_loss: 0.0459 - val_accuracy: 0.9778\n",
            "Epoch 218/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0542 - accuracy: 0.9889 - val_loss: 0.0424 - val_accuracy: 0.9778\n",
            "Epoch 219/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0578 - accuracy: 0.9667 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 220/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0574 - accuracy: 0.9778 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 221/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0531 - accuracy: 0.9778 - val_loss: 0.0434 - val_accuracy: 0.9778\n",
            "Epoch 222/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0627 - accuracy: 0.9778 - val_loss: 0.0427 - val_accuracy: 0.9778\n",
            "Epoch 223/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0631 - accuracy: 0.9667 - val_loss: 0.0421 - val_accuracy: 0.9778\n",
            "Epoch 224/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0671 - accuracy: 0.9778 - val_loss: 0.0428 - val_accuracy: 0.9778\n",
            "Epoch 225/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0467 - accuracy: 0.9778 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 226/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0564 - accuracy: 0.9778 - val_loss: 0.0429 - val_accuracy: 0.9778\n",
            "Epoch 227/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0506 - accuracy: 0.9778 - val_loss: 0.0427 - val_accuracy: 0.9778\n",
            "Epoch 228/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0655 - accuracy: 0.9778 - val_loss: 0.0425 - val_accuracy: 0.9778\n",
            "Epoch 229/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0534 - accuracy: 0.9778 - val_loss: 0.0429 - val_accuracy: 0.9778\n",
            "Epoch 230/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0670 - accuracy: 0.9667 - val_loss: 0.0507 - val_accuracy: 0.9778\n",
            "Epoch 231/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0706 - accuracy: 0.9778 - val_loss: 0.0503 - val_accuracy: 0.9778\n",
            "Epoch 232/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0677 - accuracy: 0.9667 - val_loss: 0.0475 - val_accuracy: 0.9778\n",
            "Epoch 233/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0609 - accuracy: 0.9778 - val_loss: 0.0435 - val_accuracy: 0.9778\n",
            "Epoch 234/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0558 - accuracy: 0.9778 - val_loss: 0.0424 - val_accuracy: 0.9778\n",
            "Epoch 235/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0466 - accuracy: 0.9778 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 236/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0594 - accuracy: 0.9667 - val_loss: 0.0421 - val_accuracy: 0.9778\n",
            "Epoch 237/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0478 - accuracy: 0.9778 - val_loss: 0.0430 - val_accuracy: 0.9778\n",
            "Epoch 238/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0540 - accuracy: 0.9778 - val_loss: 0.0425 - val_accuracy: 0.9778\n",
            "Epoch 239/500\n",
            "12/12 [==============================] - 4s 348ms/step - loss: 0.0475 - accuracy: 0.9778 - val_loss: 0.0442 - val_accuracy: 0.9778\n",
            "Epoch 240/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0726 - accuracy: 0.9778 - val_loss: 0.0436 - val_accuracy: 0.9778\n",
            "Epoch 241/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0446 - accuracy: 0.9778 - val_loss: 0.0435 - val_accuracy: 0.9778\n",
            "Epoch 242/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0507 - accuracy: 0.9667 - val_loss: 0.0466 - val_accuracy: 0.9778\n",
            "Epoch 243/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0748 - accuracy: 0.9667 - val_loss: 0.0422 - val_accuracy: 0.9778\n",
            "Epoch 244/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0775 - accuracy: 0.9667 - val_loss: 0.0660 - val_accuracy: 0.9778\n",
            "Epoch 245/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0664 - accuracy: 0.9778 - val_loss: 0.0463 - val_accuracy: 0.9778\n",
            "Epoch 246/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0652 - accuracy: 0.9667 - val_loss: 0.0448 - val_accuracy: 0.9778\n",
            "Epoch 247/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0589 - accuracy: 0.9667 - val_loss: 0.0431 - val_accuracy: 0.9778\n",
            "Epoch 248/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0610 - accuracy: 0.9667 - val_loss: 0.0443 - val_accuracy: 0.9778\n",
            "Epoch 249/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0552 - accuracy: 0.9667 - val_loss: 0.0416 - val_accuracy: 0.9778\n",
            "Epoch 250/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0490 - accuracy: 0.9778 - val_loss: 0.0432 - val_accuracy: 0.9778\n",
            "Epoch 251/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0585 - accuracy: 0.9556 - val_loss: 0.0420 - val_accuracy: 0.9778\n",
            "Epoch 252/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0614 - accuracy: 0.9667 - val_loss: 0.0424 - val_accuracy: 0.9778\n",
            "Epoch 253/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0580 - accuracy: 0.9667 - val_loss: 0.0412 - val_accuracy: 0.9778\n",
            "Epoch 254/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0581 - accuracy: 0.9444 - val_loss: 0.0422 - val_accuracy: 0.9778\n",
            "Epoch 255/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0448 - accuracy: 0.9778 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 256/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0554 - accuracy: 0.9778 - val_loss: 0.0416 - val_accuracy: 0.9778\n",
            "Epoch 257/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0635 - accuracy: 0.9778 - val_loss: 0.0432 - val_accuracy: 0.9778\n",
            "Epoch 258/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0491 - accuracy: 0.9778 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 259/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0564 - accuracy: 0.9778 - val_loss: 0.0416 - val_accuracy: 0.9778\n",
            "Epoch 260/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0579 - accuracy: 0.9667 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 261/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0584 - accuracy: 0.9667 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 262/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0480 - accuracy: 0.9778 - val_loss: 0.0428 - val_accuracy: 0.9778\n",
            "Epoch 263/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0642 - accuracy: 0.9778 - val_loss: 0.0402 - val_accuracy: 0.9778\n",
            "Epoch 264/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0400 - accuracy: 0.9778 - val_loss: 0.0408 - val_accuracy: 0.9778\n",
            "Epoch 265/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0516 - accuracy: 0.9778 - val_loss: 0.0405 - val_accuracy: 0.9778\n",
            "Epoch 266/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0556 - accuracy: 0.9778 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 267/500\n",
            "12/12 [==============================] - 3s 234ms/step - loss: 0.0467 - accuracy: 0.9778 - val_loss: 0.0406 - val_accuracy: 0.9778\n",
            "Epoch 268/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0470 - accuracy: 0.9778 - val_loss: 0.0401 - val_accuracy: 0.9778\n",
            "Epoch 269/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0594 - accuracy: 0.9667 - val_loss: 0.0397 - val_accuracy: 0.9778\n",
            "Epoch 270/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0528 - accuracy: 0.9667 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 271/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0455 - accuracy: 0.9778 - val_loss: 0.0398 - val_accuracy: 0.9778\n",
            "Epoch 272/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0750 - accuracy: 0.9556 - val_loss: 0.0406 - val_accuracy: 0.9778\n",
            "Epoch 273/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0652 - accuracy: 0.9667 - val_loss: 0.0402 - val_accuracy: 0.9778\n",
            "Epoch 274/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0671 - accuracy: 0.9667 - val_loss: 0.0407 - val_accuracy: 0.9778\n",
            "Epoch 275/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0505 - accuracy: 0.9778 - val_loss: 0.0409 - val_accuracy: 0.9778\n",
            "Epoch 276/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0397 - accuracy: 1.0000 - val_loss: 0.0403 - val_accuracy: 0.9778\n",
            "Epoch 277/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0582 - accuracy: 0.9667 - val_loss: 0.0404 - val_accuracy: 0.9778\n",
            "Epoch 278/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0529 - accuracy: 0.9556 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 279/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0367 - accuracy: 0.9889 - val_loss: 0.0399 - val_accuracy: 0.9778\n",
            "Epoch 280/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0529 - accuracy: 0.9667 - val_loss: 0.0397 - val_accuracy: 0.9778\n",
            "Epoch 281/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0322 - accuracy: 0.9889 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 282/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0360 - accuracy: 0.9778 - val_loss: 0.0392 - val_accuracy: 0.9778\n",
            "Epoch 283/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0490 - accuracy: 0.9889 - val_loss: 0.0393 - val_accuracy: 0.9778\n",
            "Epoch 284/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0531 - accuracy: 0.9889 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 285/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0595 - accuracy: 0.9889 - val_loss: 0.0423 - val_accuracy: 0.9778\n",
            "Epoch 286/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0694 - accuracy: 0.9778 - val_loss: 0.0394 - val_accuracy: 0.9778\n",
            "Epoch 287/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0465 - accuracy: 0.9778 - val_loss: 0.0406 - val_accuracy: 0.9778\n",
            "Epoch 288/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0388 - accuracy: 0.9667 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 289/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0462 - accuracy: 0.9889 - val_loss: 0.0392 - val_accuracy: 0.9778\n",
            "Epoch 290/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0665 - accuracy: 0.9444 - val_loss: 0.0391 - val_accuracy: 0.9778\n",
            "Epoch 291/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0521 - accuracy: 0.9778 - val_loss: 0.0426 - val_accuracy: 0.9778\n",
            "Epoch 292/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0537 - accuracy: 0.9667 - val_loss: 0.0405 - val_accuracy: 0.9778\n",
            "Epoch 293/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0428 - accuracy: 0.9778 - val_loss: 0.0414 - val_accuracy: 0.9778\n",
            "Epoch 294/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0555 - accuracy: 0.9778 - val_loss: 0.0438 - val_accuracy: 0.9778\n",
            "Epoch 295/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0527 - accuracy: 0.9667 - val_loss: 0.0424 - val_accuracy: 0.9778\n",
            "Epoch 296/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0517 - accuracy: 0.9667 - val_loss: 0.0420 - val_accuracy: 0.9778\n",
            "Epoch 297/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0521 - accuracy: 0.9667 - val_loss: 0.0424 - val_accuracy: 0.9778\n",
            "Epoch 298/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0527 - accuracy: 0.9556 - val_loss: 0.0407 - val_accuracy: 0.9778\n",
            "Epoch 299/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0468 - accuracy: 0.9778 - val_loss: 0.0405 - val_accuracy: 0.9778\n",
            "Epoch 300/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0519 - accuracy: 0.9778 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 301/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0529 - accuracy: 0.9778 - val_loss: 0.0394 - val_accuracy: 0.9778\n",
            "Epoch 302/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0340 - accuracy: 0.9889 - val_loss: 0.0399 - val_accuracy: 0.9778\n",
            "Epoch 303/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0515 - accuracy: 0.9778 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 304/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0514 - accuracy: 0.9667 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 305/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0524 - accuracy: 0.9778 - val_loss: 0.0397 - val_accuracy: 0.9778\n",
            "Epoch 306/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0496 - accuracy: 0.9778 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 307/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0395 - accuracy: 0.9778 - val_loss: 0.0402 - val_accuracy: 0.9778\n",
            "Epoch 308/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0617 - accuracy: 0.9667 - val_loss: 0.0403 - val_accuracy: 0.9778\n",
            "Epoch 309/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0395 - accuracy: 0.9778 - val_loss: 0.0393 - val_accuracy: 0.9778\n",
            "Epoch 310/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0480 - accuracy: 0.9778 - val_loss: 0.0391 - val_accuracy: 0.9778\n",
            "Epoch 311/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0527 - accuracy: 0.9778 - val_loss: 0.0397 - val_accuracy: 0.9778\n",
            "Epoch 312/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0510 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 313/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0414 - accuracy: 0.9778 - val_loss: 0.0399 - val_accuracy: 0.9778\n",
            "Epoch 314/500\n",
            "12/12 [==============================] - 3s 235ms/step - loss: 0.0499 - accuracy: 0.9667 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 315/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0394 - accuracy: 0.9778 - val_loss: 0.0387 - val_accuracy: 0.9778\n",
            "Epoch 316/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0464 - accuracy: 0.9778 - val_loss: 0.0387 - val_accuracy: 0.9778\n",
            "Epoch 317/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0617 - accuracy: 0.9778 - val_loss: 0.0401 - val_accuracy: 0.9778\n",
            "Epoch 318/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0506 - accuracy: 0.9667 - val_loss: 0.0401 - val_accuracy: 0.9778\n",
            "Epoch 319/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0411 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 320/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0575 - accuracy: 0.9778 - val_loss: 0.0390 - val_accuracy: 0.9778\n",
            "Epoch 321/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0463 - accuracy: 0.9778 - val_loss: 0.0399 - val_accuracy: 0.9778\n",
            "Epoch 322/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0577 - accuracy: 0.9778 - val_loss: 0.0406 - val_accuracy: 0.9778\n",
            "Epoch 323/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0457 - accuracy: 0.9778 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 324/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0485 - accuracy: 0.9778 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 325/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0479 - accuracy: 0.9778 - val_loss: 0.0392 - val_accuracy: 0.9778\n",
            "Epoch 326/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0443 - accuracy: 0.9778 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 327/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0445 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 328/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0541 - accuracy: 0.9667 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 329/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0494 - accuracy: 0.9778 - val_loss: 0.0394 - val_accuracy: 0.9778\n",
            "Epoch 330/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0651 - accuracy: 0.9778 - val_loss: 0.0393 - val_accuracy: 0.9778\n",
            "Epoch 331/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0637 - accuracy: 0.9667 - val_loss: 0.0393 - val_accuracy: 0.9778\n",
            "Epoch 332/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0394 - accuracy: 0.9667 - val_loss: 0.0389 - val_accuracy: 0.9778\n",
            "Epoch 333/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0473 - accuracy: 0.9778 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 334/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0437 - accuracy: 0.9778 - val_loss: 0.0385 - val_accuracy: 0.9778\n",
            "Epoch 335/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0398 - accuracy: 0.9778 - val_loss: 0.0389 - val_accuracy: 0.9778\n",
            "Epoch 336/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0367 - accuracy: 0.9889 - val_loss: 0.0390 - val_accuracy: 0.9778\n",
            "Epoch 337/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0480 - accuracy: 0.9778 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 338/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0526 - accuracy: 0.9556 - val_loss: 0.0385 - val_accuracy: 0.9778\n",
            "Epoch 339/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0599 - accuracy: 0.9667 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 340/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0403 - accuracy: 0.9667 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 341/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0416 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 342/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0503 - accuracy: 0.9667 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 343/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0323 - accuracy: 0.9778 - val_loss: 0.0381 - val_accuracy: 0.9778\n",
            "Epoch 344/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0547 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 345/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0486 - accuracy: 0.9778 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 346/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0466 - accuracy: 0.9778 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 347/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0483 - accuracy: 0.9667 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 348/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0327 - accuracy: 0.9667 - val_loss: 0.0391 - val_accuracy: 0.9778\n",
            "Epoch 349/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0375 - accuracy: 0.9889 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 350/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0401 - accuracy: 0.9667 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 351/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0392 - accuracy: 0.9667 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 352/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0420 - accuracy: 0.9778 - val_loss: 0.0413 - val_accuracy: 0.9778\n",
            "Epoch 353/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0560 - accuracy: 0.9556 - val_loss: 0.0411 - val_accuracy: 0.9778\n",
            "Epoch 354/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0700 - accuracy: 0.9556 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 355/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0307 - accuracy: 0.9889 - val_loss: 0.0387 - val_accuracy: 0.9778\n",
            "Epoch 356/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0374 - accuracy: 0.9778 - val_loss: 0.0393 - val_accuracy: 0.9778\n",
            "Epoch 357/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0464 - accuracy: 0.9778 - val_loss: 0.0385 - val_accuracy: 0.9778\n",
            "Epoch 358/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0522 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 359/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0451 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 360/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0404 - accuracy: 0.9778 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 361/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0464 - accuracy: 0.9889 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 362/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0398 - accuracy: 0.9667 - val_loss: 0.0390 - val_accuracy: 0.9778\n",
            "Epoch 363/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0494 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 364/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0390 - accuracy: 0.9667 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 365/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0480 - accuracy: 0.9889 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 366/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0483 - accuracy: 0.9667 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 367/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0443 - accuracy: 0.9667 - val_loss: 0.0419 - val_accuracy: 0.9667\n",
            "Epoch 368/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0545 - accuracy: 0.9667 - val_loss: 0.0409 - val_accuracy: 0.9778\n",
            "Epoch 369/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.0552 - accuracy: 0.9778 - val_loss: 0.0397 - val_accuracy: 0.9778\n",
            "Epoch 370/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0466 - accuracy: 0.9889 - val_loss: 0.0390 - val_accuracy: 0.9778\n",
            "Epoch 371/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0433 - accuracy: 0.9778 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 372/500\n",
            "12/12 [==============================] - 4s 351ms/step - loss: 0.0365 - accuracy: 0.9889 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 373/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0438 - accuracy: 0.9667 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 374/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0590 - accuracy: 0.9667 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 375/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0467 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 376/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0457 - accuracy: 0.9778 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 377/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0412 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 378/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0374 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 379/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0434 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 380/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0505 - accuracy: 0.9889 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 381/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.0438 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 382/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0497 - accuracy: 0.9778 - val_loss: 0.0385 - val_accuracy: 0.9778\n",
            "Epoch 383/500\n",
            "12/12 [==============================] - 4s 353ms/step - loss: 0.0363 - accuracy: 0.9778 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 384/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0497 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 385/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0388 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 386/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0545 - accuracy: 0.9778 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 387/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0480 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 388/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0464 - accuracy: 0.9778 - val_loss: 0.0381 - val_accuracy: 0.9778\n",
            "Epoch 389/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0430 - accuracy: 0.9667 - val_loss: 0.0405 - val_accuracy: 0.9667\n",
            "Epoch 390/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0480 - accuracy: 0.9667 - val_loss: 0.0404 - val_accuracy: 0.9778\n",
            "Epoch 391/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0496 - accuracy: 0.9667 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 392/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0491 - accuracy: 0.9889 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 393/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0420 - accuracy: 0.9778 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 394/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0536 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 395/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0431 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 396/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0409 - accuracy: 0.9667 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 397/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0397 - accuracy: 0.9778 - val_loss: 0.0381 - val_accuracy: 0.9778\n",
            "Epoch 398/500\n",
            "12/12 [==============================] - 3s 241ms/step - loss: 0.0446 - accuracy: 0.9889 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 399/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0386 - accuracy: 0.9889 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 400/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0490 - accuracy: 0.9778 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 401/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0394 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 402/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0567 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 403/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0467 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 404/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0429 - accuracy: 0.9667 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 405/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0465 - accuracy: 0.9667 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 406/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0572 - accuracy: 0.9667 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 407/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0516 - accuracy: 0.9778 - val_loss: 0.0416 - val_accuracy: 0.9778\n",
            "Epoch 408/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0633 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 409/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0500 - accuracy: 0.9667 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 410/500\n",
            "12/12 [==============================] - 4s 352ms/step - loss: 0.0423 - accuracy: 0.9778 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 411/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0448 - accuracy: 0.9667 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 412/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0420 - accuracy: 0.9889 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 413/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0497 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 414/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0471 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 415/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0392 - accuracy: 0.9778 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 416/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0377 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 417/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0423 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 418/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0407 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 419/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0571 - accuracy: 0.9667 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 420/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0428 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 421/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0449 - accuracy: 0.9778 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 422/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0475 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 423/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0452 - accuracy: 0.9778 - val_loss: 0.0415 - val_accuracy: 0.9778\n",
            "Epoch 424/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0462 - accuracy: 0.9778 - val_loss: 0.0415 - val_accuracy: 0.9778\n",
            "Epoch 425/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0552 - accuracy: 0.9778 - val_loss: 0.0395 - val_accuracy: 0.9778\n",
            "Epoch 426/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0464 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 427/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0453 - accuracy: 0.9667 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 428/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0410 - accuracy: 0.9667 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 429/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0426 - accuracy: 0.9778 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 430/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0494 - accuracy: 0.9778 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 431/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0449 - accuracy: 0.9778 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 432/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0446 - accuracy: 0.9667 - val_loss: 0.0373 - val_accuracy: 0.9778\n",
            "Epoch 433/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0499 - accuracy: 0.9778 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 434/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0421 - accuracy: 0.9556 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 435/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0434 - accuracy: 0.9778 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 436/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0504 - accuracy: 0.9778 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 437/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0372 - accuracy: 0.9778 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 438/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0426 - accuracy: 0.9778 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 439/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0543 - accuracy: 0.9556 - val_loss: 0.0406 - val_accuracy: 0.9667\n",
            "Epoch 440/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0548 - accuracy: 0.9667 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 441/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0477 - accuracy: 0.9778 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 442/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0374 - accuracy: 0.9667 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 443/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0460 - accuracy: 0.9667 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 444/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0509 - accuracy: 0.9778 - val_loss: 0.0373 - val_accuracy: 0.9778\n",
            "Epoch 445/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0465 - accuracy: 0.9667 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 446/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0419 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 447/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0416 - accuracy: 0.9778 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 448/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0443 - accuracy: 0.9778 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 449/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0363 - accuracy: 0.9889 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 450/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0548 - accuracy: 0.9778 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 451/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0525 - accuracy: 0.9667 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 452/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0536 - accuracy: 0.9778 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 453/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0490 - accuracy: 0.9667 - val_loss: 0.0463 - val_accuracy: 0.9667\n",
            "Epoch 454/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0581 - accuracy: 0.9778 - val_loss: 0.0423 - val_accuracy: 0.9667\n",
            "Epoch 455/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0377 - accuracy: 0.9667 - val_loss: 0.0394 - val_accuracy: 0.9778\n",
            "Epoch 456/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0378 - accuracy: 0.9778 - val_loss: 0.0404 - val_accuracy: 0.9778\n",
            "Epoch 457/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0600 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 458/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0554 - accuracy: 0.9889 - val_loss: 0.0419 - val_accuracy: 0.9667\n",
            "Epoch 459/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0463 - accuracy: 0.9889 - val_loss: 0.0421 - val_accuracy: 0.9667\n",
            "Epoch 460/500\n",
            "12/12 [==============================] - 4s 350ms/step - loss: 0.0493 - accuracy: 0.9667 - val_loss: 0.0403 - val_accuracy: 0.9778\n",
            "Epoch 461/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0512 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 462/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0468 - accuracy: 0.9778 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 463/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0464 - accuracy: 0.9556 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 464/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0456 - accuracy: 0.9778 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 465/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0414 - accuracy: 0.9889 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 466/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0463 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 467/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0473 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 468/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0366 - accuracy: 0.9889 - val_loss: 0.0375 - val_accuracy: 0.9778\n",
            "Epoch 469/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0443 - accuracy: 0.9889 - val_loss: 0.0381 - val_accuracy: 0.9778\n",
            "Epoch 470/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0493 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 471/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0416 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 472/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0373 - accuracy: 0.9778 - val_loss: 0.0372 - val_accuracy: 0.9778\n",
            "Epoch 473/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0344 - accuracy: 0.9889 - val_loss: 0.0374 - val_accuracy: 0.9778\n",
            "Epoch 474/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0592 - accuracy: 0.9778 - val_loss: 0.0376 - val_accuracy: 0.9778\n",
            "Epoch 475/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0426 - accuracy: 0.9667 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 476/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0639 - accuracy: 0.9667 - val_loss: 0.0378 - val_accuracy: 0.9778\n",
            "Epoch 477/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0424 - accuracy: 0.9889 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 478/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0688 - accuracy: 0.9556 - val_loss: 0.0396 - val_accuracy: 0.9778\n",
            "Epoch 479/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0772 - accuracy: 0.9444 - val_loss: 0.0472 - val_accuracy: 0.9667\n",
            "Epoch 480/500\n",
            "12/12 [==============================] - 4s 352ms/step - loss: 0.0452 - accuracy: 0.9889 - val_loss: 0.0449 - val_accuracy: 0.9667\n",
            "Epoch 481/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0663 - accuracy: 0.9667 - val_loss: 0.0400 - val_accuracy: 0.9778\n",
            "Epoch 482/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0455 - accuracy: 0.9778 - val_loss: 0.0394 - val_accuracy: 0.9778\n",
            "Epoch 483/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0586 - accuracy: 0.9667 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 484/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0409 - accuracy: 0.9778 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 485/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0455 - accuracy: 0.9889 - val_loss: 0.0388 - val_accuracy: 0.9778\n",
            "Epoch 486/500\n",
            "12/12 [==============================] - 3s 240ms/step - loss: 0.0435 - accuracy: 0.9778 - val_loss: 0.0386 - val_accuracy: 0.9778\n",
            "Epoch 487/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0459 - accuracy: 0.9778 - val_loss: 0.0379 - val_accuracy: 0.9778\n",
            "Epoch 488/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0444 - accuracy: 0.9778 - val_loss: 0.0384 - val_accuracy: 0.9778\n",
            "Epoch 489/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0416 - accuracy: 0.9778 - val_loss: 0.0383 - val_accuracy: 0.9778\n",
            "Epoch 490/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0427 - accuracy: 0.9778 - val_loss: 0.0380 - val_accuracy: 0.9778\n",
            "Epoch 491/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0529 - accuracy: 0.9667 - val_loss: 0.0387 - val_accuracy: 0.9778\n",
            "Epoch 492/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0445 - accuracy: 0.9667 - val_loss: 0.0392 - val_accuracy: 0.9778\n",
            "Epoch 493/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0414 - accuracy: 0.9778 - val_loss: 0.0409 - val_accuracy: 0.9778\n",
            "Epoch 494/500\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0579 - accuracy: 0.9556 - val_loss: 0.0420 - val_accuracy: 0.9667\n",
            "Epoch 495/500\n",
            "12/12 [==============================] - 3s 236ms/step - loss: 0.0518 - accuracy: 0.9667 - val_loss: 0.0387 - val_accuracy: 0.9778\n",
            "Epoch 496/500\n",
            "12/12 [==============================] - 4s 351ms/step - loss: 0.0385 - accuracy: 0.9778 - val_loss: 0.0390 - val_accuracy: 0.9778\n",
            "Epoch 497/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0472 - accuracy: 0.9667 - val_loss: 0.0382 - val_accuracy: 0.9778\n",
            "Epoch 498/500\n",
            "12/12 [==============================] - 3s 239ms/step - loss: 0.0478 - accuracy: 0.9778 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 499/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0398 - accuracy: 0.9667 - val_loss: 0.0377 - val_accuracy: 0.9778\n",
            "Epoch 500/500\n",
            "12/12 [==============================] - 3s 237ms/step - loss: 0.0433 - accuracy: 0.9889 - val_loss: 0.0378 - val_accuracy: 0.9778\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!ls"
      ],
      "metadata": {
        "id": "Ju76yb4Fcutv",
        "outputId": "711aa8cb-b1f7-4a72-e185-91e8f8888e27",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " dataset\t\t      ResNet101V2_COVID_19.ipynb\n",
            "'\\dataset1'\t\t      ResNet152_COVID_19.ipynb\n",
            " DenseNet121_COVID_19.ipynb   ResNet152V2_COVID_19.ipynb\n",
            " DenseNet169_COVID_19.ipynb   ResNet50V2_COVID_19.ipynb\n",
            " DenseNet201_COVID_19.ipynb   train_model.ipynb\n",
            " log.csv\t\t      VGG16_COVID_19.ipynb\n",
            " output\t\t\t      VGG19_COVID_19.ipynb\n",
            " README.md\t\t      Wazir_first_model.png\n",
            " ResNet101_COVID_19.ipynb     Xception_COVID_19.ipynb\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import files\n",
        "files.download('log.csv')"
      ],
      "metadata": {
        "id": "Ucfz-wlld7RG",
        "outputId": "7e7692e5-5b67-42c0-b9bf-554d5f665685",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        }
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "download(\"download_af74921a-829e-44ad-862d-f8a9d8734d89\", \"log.csv\", 41448)"
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "fIgAEIEijGni"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}