Range_Doppler_LSTM_v2_datalarla_dropout03.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mertege/FMCW-Data-Classification-/blob/main/Range_Doppler_LSTM_v2_datalarla_dropout03.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kN4ixvH39HW6"
      },
      "outputs": [],
      "source": [
        "#dropout selected as 0.3, \n",
        "#[0.9823529362678528, 0.970588219165802, 0.9823529243469238, 0.9764705777168274, 0.9647058606147766]\n",
        "#[0.9789437689969607, 0.9654841563514089, 0.9796508230180756, 0.9736615994756604, 0.9603545628048348]\n",
        "#Mean test accuracy is 0.98, mean test f1 score is 0.97, max test accuracy is 0.98, max test f1 score is 0.98, min test accuracy is 0.96, min test f1 score is 0.96, std of test accuracy is 0.01, std of test f1 score is 0.01\n",
        "#Time elapsed through all process: 8976.75, sec"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "lzJwN6GfkN8Z"
      },
      "outputs": [],
      "source": [
        "from keras.models import Sequential\n",
        "from tensorflow.keras import layers\n",
        "from keras.models import Model\n",
        "from keras.layers import Activation, Dropout, Flatten, Dense, BatchNormalization, Normalization, Input, Conv2D, MaxPooling2D, Concatenate, GRU, LSTM, GRU, TimeDistributed, Bidirectional\n",
        "import tensorflow as tf\n",
        "import scipy.io\n",
        "import numpy as np\n",
        "import cv2\n",
        "import matplotlib.pyplot as plt\n",
        "import random\n",
        "from numpy.random import seed\n",
        "from sklearn.model_selection import KFold, StratifiedKFold\n",
        "import time\n",
        "from sklearn.metrics import precision_recall_fscore_support\n",
        "from keras.callbacks import EarlyStopping\n",
        "from numpy import array\n",
        "from numpy import argmax\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from keras import backend as K \n",
        "import gc\n",
        "import os"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8FERVnWrkrwE",
        "outputId": "5dc67af2-e5b7-42d2-89a2-604804013b2c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "o19JHieQ81cS"
      },
      "outputs": [],
      "source": [
        "# Get Range-Doppler data from\n",
        "range_doppler_fast_resized = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_fast_resizedv2.mat')\n",
        "range_doppler_fast_resized = range_doppler_fast_resized['range_doppler_fast_resized']\n",
        "range_doppler_fast_resized = np.transpose(range_doppler_fast_resized, (2, 0, 1))\n",
        "# range_doppler_fast_resized = np.delete(range_doppler_fast_resized,(49), axis=0) # 50th row is deleted since there is no 50th row in spectrogram fast data.\n",
        "range_doppler_fast_label = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_fast_label.mat')\n",
        "range_doppler_fast_label = range_doppler_fast_label['range_doppler_fast_label']  \n",
        "\n",
        "range_doppler_slow_resized = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_slow_resizedv2.mat')\n",
        "range_doppler_slow_resized = range_doppler_slow_resized['range_doppler_slow_resized']\n",
        "range_doppler_slow_resized = np.transpose(range_doppler_slow_resized, (2, 0, 1))\n",
        "range_doppler_slow_label = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_slow_label.mat')\n",
        "range_doppler_slow_label = range_doppler_slow_label['range_doppler_slow_label']  \n",
        "\n",
        "range_doppler_slow_pocket_resized = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_slow_pocket_resizedv2.mat')\n",
        "range_doppler_slow_pocket_resized = range_doppler_slow_pocket_resized['range_doppler_slow_pocket_resized']\n",
        "range_doppler_slow_pocket_resized = np.transpose(range_doppler_slow_pocket_resized, (2, 0, 1))\n",
        "range_doppler_pocket_label = scipy.io.loadmat('/content/drive/MyDrive/data/range_doppler_pocket_label.mat')\n",
        "range_doppler_pocket_label = range_doppler_pocket_label['range_doppler_pocket_label']  \n",
        "# Get Range-Doppler data from\n",
        "spectrogram_fast_resized = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_fast_resized.mat')\n",
        "spectrogram_fast_resized = spectrogram_fast_resized['spectrogram_fast_resized']\n",
        "spectrogram_fast_resized = np.transpose(spectrogram_fast_resized, (2, 0, 1))\n",
        "spectrogram_fast_label = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_fast_label.mat')\n",
        "spectrogram_fast_label = spectrogram_fast_label['spectrogram_fast_label']  \n",
        "\n",
        "spectrogram_slow_resized = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_slow_resized.mat')\n",
        "spectrogram_slow_resized = spectrogram_slow_resized['spectrogram_slow_resized']\n",
        "spectrogram_slow_resized = np.transpose(spectrogram_slow_resized, (2, 0, 1))\n",
        "spectrogram_slow_label = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_slow_label.mat')\n",
        "spectrogram_slow_label = spectrogram_slow_label['spectrogram_slow_label']  \n",
        "\n",
        "spectrogram_slow_pocket_resized = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_slow_pocket_resized.mat')\n",
        "spectrogram_slow_pocket_resized = spectrogram_slow_pocket_resized['spectrogram_slow_pocket_resized']\n",
        "spectrogram_slow_pocket_resized = np.transpose(spectrogram_slow_pocket_resized, (2, 0, 1))\n",
        "spectrogram_slow_pocket_label = scipy.io.loadmat('/content/drive/MyDrive/data/spectrogram_slow_pocket_label.mat')\n",
        "spectrogram_slow_pocket_label = spectrogram_slow_pocket_label['spectrogram_slow_pocket_label']  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "wFp_PErnMLN2"
      },
      "outputs": [],
      "source": [
        "# Concat range-doppler data\n",
        "range_doppler_concat = np.concatenate((range_doppler_fast_resized,range_doppler_slow_resized),axis=0)\n",
        "range_doppler_concat = np.concatenate((range_doppler_concat,range_doppler_slow_pocket_resized),axis=0)\n",
        "range_doppler_concat = range_doppler_concat[:,:,:,np.newaxis] \n",
        "range_doppler_concat_label = np.zeros((range_doppler_concat.shape[0],1))\n",
        "range_doppler_concat_label[:range_doppler_fast_resized.shape[0],:] = 1\n",
        "# Shuffle concat range doppler\n",
        "shuffle_indx = random.sample(range(0, range_doppler_concat.shape[0]), range_doppler_concat.shape[0]) # split validation data\n",
        "range_doppler_concat_shuffle = range_doppler_concat[shuffle_indx,:,:,:]\n",
        "range_doppler_concat_label_shuffle = range_doppler_concat_label[shuffle_indx,:]\n",
        "# Concat range-doppler data\n",
        "spectrogram_concat = np.concatenate((spectrogram_fast_resized,spectrogram_slow_resized),axis=0)\n",
        "spectrogram_concat = np.concatenate((spectrogram_concat,spectrogram_slow_pocket_resized),axis=0)\n",
        "spectrogram_concat = spectrogram_concat[:,:,:,np.newaxis] \n",
        "spectrogram_concat_label = np.zeros((spectrogram_concat.shape[0],1))\n",
        "spectrogram_concat_label[:spectrogram_fast_resized.shape[0],:] = 1\n",
        "# Shuffle concat range doppler\n",
        "# shuffle_indx = random.sample(range(0, spectrogram_concat.shape[0]), spectrogram_concat.shape[0]) # split validation data\n",
        "spectrogram_concat_shuffle = spectrogram_concat[shuffle_indx,:,:,:]\n",
        "spectrogram_concat_label_shuffle = spectrogram_concat_label[shuffle_indx,:]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "_1aBNkjoLCvI"
      },
      "outputs": [],
      "source": [
        "# ---------------- Augmente and shuffle (train and test) data data ----------------\n",
        "data_augmentation = tf.keras.Sequential([\n",
        "  layers.RandomFlip(\"horizontal_and_vertical\"),\n",
        "  layers.RandomRotation(0.1),\n",
        "])\n",
        "\n",
        "\n",
        "def mixup_augmentation(images,range_doppler_training_data, labels, repeat_of_mixup, alpha=0.2):\n",
        "    batch_size = images.shape[0]\n",
        "    concat_images = np.zeros((batch_size*(repeat_of_mixup+1),images.shape[1],images.shape[2],images.shape[3]))\n",
        "    concat_images_range_doppler = np.zeros((batch_size*(repeat_of_mixup+1),range_doppler_training_data.shape[1],\\\n",
        "                                            range_doppler_training_data.shape[2],range_doppler_training_data.shape[3]))\n",
        "    concat_label = np.zeros((batch_size*(repeat_of_mixup+1),labels.shape[1]))\n",
        "    for ii in range(repeat_of_mixup):\n",
        "      # shuffle train dataset\n",
        "      shuffle_indx_1 = random.sample(range(0, images.shape[0]), images.shape[0]) # split validation data\n",
        "      images_shuffled_1 = images[shuffle_indx_1,:,:,:]\n",
        "      range_doppler_training_data_shuffled_1 = range_doppler_training_data[shuffle_indx_1,:,:,:]\n",
        "      labels_shuffled_1 = labels[shuffle_indx_1,:]\n",
        "\n",
        "      shuffle_indx_2 = random.sample(range(0, images.shape[0]), images.shape[0]) # split validation data\n",
        "      images_shuffled_2 = images[shuffle_indx_2,:,:,:]\n",
        "      range_doppler_training_data_shuffled_2 = range_doppler_training_data[shuffle_indx_2,:,:,:]\n",
        "      labels_shuffled_2 = labels[shuffle_indx_2,:]\n",
        "\n",
        "      # Sample lambda and reshape it to do the mixup\n",
        "      gaussian_mean = 0.2\n",
        "      gaussian_std = 0.02\n",
        "      ll = np.random.normal(gaussian_mean, gaussian_std, (batch_size,1,1,1))\n",
        "      x_l = np.reshape(ll, (batch_size,1,1,1))\n",
        "      y_l = np.reshape(ll, (batch_size,1))\n",
        "      \n",
        "      # Perform mixup on both images and labels by combining a pair of images/labels\n",
        "      images_mixup = images_shuffled_1 * x_l + images_shuffled_2 * (1 - x_l)\n",
        "      images_mixup_range_doppler = range_doppler_training_data_shuffled_1 * x_l + range_doppler_training_data_shuffled_2 * (1 - x_l)\n",
        "      labels_mixup = labels_shuffled_1 * y_l + labels_shuffled_2 * (1 - y_l)\n",
        "      concat_images[ii*batch_size:(ii+1)*batch_size,:,:,:] = images_mixup\n",
        "      concat_images_range_doppler[ii*batch_size:(ii+1)*batch_size,:,:,:] = images_mixup_range_doppler\n",
        "      concat_label[ii*batch_size:(ii+1)*batch_size,:] = labels_mixup\n",
        "\n",
        "    concat_images[repeat_of_mixup*batch_size:,:,:,:] = images\n",
        "    concat_images_range_doppler[repeat_of_mixup*batch_size:,:,:,:] = range_doppler_training_data\n",
        "    concat_label[repeat_of_mixup*batch_size:,:] = labels\n",
        "    return (concat_images,concat_images_range_doppler, concat_label)\n",
        "def split_and_augmentation_of_training(spectrogram_concat_shuffle_train,range_doppler_concat_shuffle_train,range_doppler_concat_label_shuffle_train,\\\n",
        "                                       repeat_of_mixup, augmentation_enable):\n",
        "  # ---------------- Parameters ----------------\n",
        "  repeat_of_augmentation_for_fast = 1\n",
        "  repeat_of_augmentation_for_slow = np.floor(repeat_of_augmentation_for_fast/2)\n",
        "  repeat_of_augmentation_for_slow = int(repeat_of_augmentation_for_slow)\n",
        "  # size_of_validation = 30\n",
        "  alpha = 0.2\n",
        "  dummy_label = np.zeros((spectrogram_concat_shuffle_train.shape[0],1))\n",
        "  for randomlist_for_train_indx, randomlist_for_validation_indx in kfold.split(spectrogram_concat_shuffle_train,dummy_label):   \n",
        "    randomlist_for_validation_indx\n",
        "  # Split validation\n",
        "  # randomlist_for_validation_indx = random.sample(range(0, range_doppler_concat_shuffle_train.shape[0]), size_of_validation) # split validation data\n",
        "  # randomlist_for_train_indx = np.delete(range(0, range_doppler_concat_shuffle_train.shape[0]), randomlist_for_validation_indx) # split training data\n",
        "  # get validation data\n",
        "  spectrogram_validation_data = spectrogram_concat_shuffle_train[randomlist_for_validation_indx,:,:,:]\n",
        "  spectrogram_validation_labels = range_doppler_concat_label_shuffle_train[randomlist_for_validation_indx,:]\n",
        "  range_doppler_validation_data = range_doppler_concat_shuffle_train[randomlist_for_validation_indx,:,:,:]\n",
        "  # get training data\n",
        "  spectrogram_training_data = spectrogram_concat_shuffle_train[randomlist_for_train_indx,:,:,:]\n",
        "  spectrogram_training_labels = spectrogram_concat_label_shuffle_train[randomlist_for_train_indx,:]\n",
        "  range_doppler_training_data = range_doppler_concat_shuffle_train[randomlist_for_train_indx,:,:,:]\n",
        "\n",
        "  # Rotate Augmentation\n",
        "  # get slow and fast indexes of training data\n",
        "  slow_indexes = np.where(spectrogram_training_labels == 0)[0]\n",
        "  fast_indexes = np.delete(range(0, spectrogram_training_labels.shape[0]), slow_indexes)  \n",
        "\n",
        "  slow_spectrograms_train = spectrogram_training_data[slow_indexes,:,:,:]\n",
        "  size_of_samples_slow = slow_spectrograms_train.shape[0]\n",
        "\n",
        "  fast_spectrograms_train = spectrogram_training_data[fast_indexes,:,:,:]  \n",
        "  size_of_samples_fast = fast_spectrograms_train.shape[0]\n",
        "\n",
        "  slow_range_train = range_doppler_training_data[slow_indexes,:,:,:]\n",
        "  fast_range_train = range_doppler_training_data[fast_indexes,:,:,:]  \n",
        "\n",
        "  if augmentation_enable == True: \n",
        "    # ---------------- Augmente Train Data for Fast ----------------\n",
        "    augmented_image_fast = np.zeros((size_of_samples_fast*repeat_of_augmentation_for_fast,fast_spectrograms_train.shape[1],fast_spectrograms_train.shape[2],1))\n",
        "    spectrograms_fast_label = np.ones((size_of_samples_fast*(repeat_of_augmentation_for_fast+1),1))\n",
        "    # augmented_image_fast = np.flip(fast_spectrograms_train,axis=2)\n",
        "    for jj in range(repeat_of_augmentation_for_fast):\n",
        "      for ii in range(size_of_samples_fast):\n",
        "        augmented_image_fast[size_of_samples_fast*jj+ii,:,:,:] = data_augmentation(fast_spectrograms_train[ii,:,:,:])\n",
        "    augmented_image_fast = np.concatenate((augmented_image_fast,fast_spectrograms_train),axis=0)   \n",
        "\n",
        "    augmented_image_fast_range = np.zeros((size_of_samples_fast*repeat_of_augmentation_for_fast,fast_range_train.shape[1],fast_range_train.shape[2],1))\n",
        "    # augmented_image_fast_range  = np.flip(fast_range_train,axis=2)\n",
        "    for jj in range(repeat_of_augmentation_for_fast):\n",
        "      for ii in range(size_of_samples_fast):\n",
        "        augmented_image_fast_range[size_of_samples_fast*jj+ii,:,:,:]  =data_augmentation(fast_range_train[ii,:,:,:])\n",
        "    augmented_image_fast_range = np.concatenate((augmented_image_fast_range,fast_range_train),axis=0)   \n",
        "\n",
        "    # ---------------- Augmente Train Data for Slow ----------------\n",
        "    augmented_image_slow = slow_spectrograms_train\n",
        "    augmented_image_slow_range = slow_range_train\n",
        "\n",
        "  else:\n",
        "    augmented_image_fast = fast_spectrograms_train\n",
        "    augmented_image_slow = slow_spectrograms_train\n",
        "    augmented_image_fast_range = fast_range_train\n",
        "    augmented_image_slow_range = slow_range_train\n",
        "    spectrograms_fast_label = np.ones((size_of_samples_fast,1))\n",
        "    \n",
        "  spectrograms_slow_label = np.zeros((size_of_samples_slow,1))\n",
        "\n",
        "  spectrogram_training_data = np.concatenate((augmented_image_fast,augmented_image_slow),axis=0)\n",
        "  range_doppler_training_data = np.concatenate((augmented_image_fast_range,augmented_image_slow_range),axis=0)\n",
        "  spectrogram_training_labels = np.concatenate((spectrograms_fast_label,spectrograms_slow_label),axis=0)\n",
        "\n",
        "  (spectrogram_augmented_image,range_doppler_augmented_image,spectrograms_label)=\\\n",
        "   mixup_augmentation(spectrogram_training_data,range_doppler_training_data, spectrogram_training_labels, repeat_of_mixup, alpha=0.2)\n",
        "\n",
        "  return (spectrogram_augmented_image,range_doppler_augmented_image,spectrograms_label,\\\n",
        "     spectrogram_validation_data,range_doppler_validation_data, spectrogram_validation_labels)\n",
        "def normalize_inputs(range_doppler_concat_shuffle, normalize_inputs_enable):\n",
        "  # ---------------- Normalize Inputs ----------------\n",
        "  if normalize_inputs_enable == True:\n",
        "    layer = Normalization(axis=None)\n",
        "    layer.adapt(range_doppler_concat_shuffle)\n",
        "    range_doppler_concat_shuffle = layer(range_doppler_concat_shuffle)\n",
        "  else:\n",
        "    range_doppler_concat_shuffle = range_doppler_concat_shuffle\n",
        "  return(range_doppler_concat_shuffle)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "8CUvxggth7ZQ"
      },
      "outputs": [],
      "source": [
        "normalize_inputs_enable = 1\n",
        "range_doppler_concat_shuffle = normalize_inputs(range_doppler_concat_shuffle, normalize_inputs_enable)\n",
        "spectrogram_concat_shuffle = normalize_inputs(spectrogram_concat_shuffle, normalize_inputs_enable)\n",
        "range_doppler_concat_shuffle = np.float32(range_doppler_concat_shuffle)\n",
        "spectrogram_concat_shuffle = np.float32(spectrogram_concat_shuffle)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "38jkSYXM4aey"
      },
      "outputs": [],
      "source": [
        "n_features = range_doppler_concat_shuffle.shape[1]\n",
        "n_steps = range_doppler_concat_shuffle.shape[2]\n",
        "range_doppler_concat_shuffle = np.transpose(range_doppler_concat_shuffle, axes = (0,2,1,3)) \n",
        "spectrogram_concat_shuffle = np.transpose(spectrogram_concat_shuffle, axes = (0,2,1,3)) "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rNpkBBIadpPo",
        "outputId": "8c704677-8152-424d-8a89-8f78e1048d20"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 5120, 32)         8832      \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 163840)            0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 163840)           655360    \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 163840)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               41943296  \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 256)              1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 256)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,608,512\n",
            "Trainable params: 42,280,320\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " model (Functional)          (None, 256)               42608512  \n",
            "                                                                 \n",
            " model_1 (Functional)        (None, 1)                 82305     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,690,817\n",
            "Trainable params: 42,362,625\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 15s 336ms/step - loss: 0.5654 - accuracy: 0.4886 - val_loss: 0.2786 - val_accuracy: 0.9259\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.4175 - accuracy: 0.5411 - val_loss: 0.2707 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3806 - accuracy: 0.5605 - val_loss: 0.1753 - val_accuracy: 1.0000\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.3609 - accuracy: 0.5662 - val_loss: 0.1948 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.3430 - accuracy: 0.5753 - val_loss: 0.2054 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.3296 - accuracy: 0.5765 - val_loss: 0.2389 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3232 - accuracy: 0.5788 - val_loss: 0.2579 - val_accuracy: 0.8889\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3062 - accuracy: 0.5833 - val_loss: 0.1530 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.2990 - accuracy: 0.5845 - val_loss: 0.2485 - val_accuracy: 0.8889\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3140 - accuracy: 0.5822 - val_loss: 0.1155 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2934 - accuracy: 0.5822 - val_loss: 0.1013 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2891 - accuracy: 0.5822 - val_loss: 0.1647 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2844 - accuracy: 0.5833 - val_loss: 0.1556 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2820 - accuracy: 0.5833 - val_loss: 0.1700 - val_accuracy: 0.9259\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2879 - accuracy: 0.5833 - val_loss: 0.1369 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2849 - accuracy: 0.5833 - val_loss: 0.1239 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2807 - accuracy: 0.5833 - val_loss: 0.1235 - val_accuracy: 0.9259\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2786 - accuracy: 0.5833 - val_loss: 0.1833 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2758 - accuracy: 0.5822 - val_loss: 0.2651 - val_accuracy: 0.8519\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2756 - accuracy: 0.5833 - val_loss: 0.1588 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2772 - accuracy: 0.5845 - val_loss: 0.1074 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2760 - accuracy: 0.5845 - val_loss: 0.1291 - val_accuracy: 0.9259\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2700 - accuracy: 0.5845 - val_loss: 0.2237 - val_accuracy: 0.9259\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2714 - accuracy: 0.5845 - val_loss: 0.0696 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2676 - accuracy: 0.5822 - val_loss: 0.0760 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2679 - accuracy: 0.5845 - val_loss: 0.1955 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2646 - accuracy: 0.5845 - val_loss: 0.0968 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2642 - accuracy: 0.5845 - val_loss: 0.0880 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2726 - accuracy: 0.5845 - val_loss: 0.1578 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2678 - accuracy: 0.5845 - val_loss: 0.2054 - val_accuracy: 0.9259\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.2686 - accuracy: 0.5845 - val_loss: 0.0778 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2552 - accuracy: 0.5845 - val_loss: 0.1018 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.2634 - accuracy: 0.5845 - val_loss: 0.1038 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2591 - accuracy: 0.5845 - val_loss: 0.1074 - val_accuracy: 0.9259\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2622 - accuracy: 0.5845 - val_loss: 0.0514 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2610 - accuracy: 0.5845 - val_loss: 0.1126 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2558 - accuracy: 0.5845 - val_loss: 0.0921 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2621 - accuracy: 0.5845 - val_loss: 0.0930 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2618 - accuracy: 0.5845 - val_loss: 0.1393 - val_accuracy: 0.9259\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.2619 - accuracy: 0.5845 - val_loss: 0.2069 - val_accuracy: 0.8889\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2584 - accuracy: 0.5845 - val_loss: 0.1055 - val_accuracy: 0.9259\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2593 - accuracy: 0.5845 - val_loss: 0.0941 - val_accuracy: 0.9630\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2606 - accuracy: 0.5845 - val_loss: 0.0926 - val_accuracy: 1.0000\n",
            "Epoch 44/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2603 - accuracy: 0.5845 - val_loss: 0.0763 - val_accuracy: 1.0000\n",
            "Epoch 45/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2582 - accuracy: 0.5845 - val_loss: 0.0739 - val_accuracy: 1.0000\n",
            "Epoch 46/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2664 - accuracy: 0.5845 - val_loss: 0.1273 - val_accuracy: 0.8889\n",
            "Epoch 47/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2546 - accuracy: 0.5845 - val_loss: 0.0907 - val_accuracy: 0.9630\n",
            "Epoch 48/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2598 - accuracy: 0.5845 - val_loss: 0.0881 - val_accuracy: 1.0000\n",
            "Epoch 49/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2592 - accuracy: 0.5845 - val_loss: 0.0749 - val_accuracy: 1.0000\n",
            "Epoch 50/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2586 - accuracy: 0.5845 - val_loss: 0.0861 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 116ms/step - loss: 0.0426 - accuracy: 1.0000\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 5120, 32)         8832      \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 163840)            0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 163840)           655360    \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 163840)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               41943296  \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 256)              1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 256)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,608,512\n",
            "Trainable params: 42,280,320\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " model (Functional)          (None, 256)               42608512  \n",
            "                                                                 \n",
            " model_1 (Functional)        (None, 1)                 82305     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,690,817\n",
            "Trainable params: 42,362,625\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 328ms/step - loss: 0.5024 - accuracy: 0.4860 - val_loss: 0.2981 - val_accuracy: 0.8889\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3846 - accuracy: 0.5431 - val_loss: 0.3917 - val_accuracy: 0.8148\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3903 - accuracy: 0.5373 - val_loss: 0.2708 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3488 - accuracy: 0.5513 - val_loss: 0.2214 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3403 - accuracy: 0.5548 - val_loss: 0.2429 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3452 - accuracy: 0.5513 - val_loss: 0.3254 - val_accuracy: 0.8519\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3205 - accuracy: 0.5571 - val_loss: 0.2047 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3138 - accuracy: 0.5629 - val_loss: 0.1732 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3014 - accuracy: 0.5618 - val_loss: 0.1840 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3116 - accuracy: 0.5618 - val_loss: 0.1590 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2986 - accuracy: 0.5641 - val_loss: 0.1422 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2966 - accuracy: 0.5629 - val_loss: 0.1507 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2822 - accuracy: 0.5641 - val_loss: 0.2173 - val_accuracy: 0.8889\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2940 - accuracy: 0.5629 - val_loss: 0.1244 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2943 - accuracy: 0.5618 - val_loss: 0.1482 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2909 - accuracy: 0.5629 - val_loss: 0.1068 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2843 - accuracy: 0.5618 - val_loss: 0.1143 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2767 - accuracy: 0.5641 - val_loss: 0.1170 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2832 - accuracy: 0.5629 - val_loss: 0.1558 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2820 - accuracy: 0.5641 - val_loss: 0.1628 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2852 - accuracy: 0.5641 - val_loss: 0.2060 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2837 - accuracy: 0.5641 - val_loss: 0.1981 - val_accuracy: 0.9259\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2826 - accuracy: 0.5641 - val_loss: 0.1967 - val_accuracy: 0.9259\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2715 - accuracy: 0.5641 - val_loss: 0.1475 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2793 - accuracy: 0.5641 - val_loss: 0.1326 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2798 - accuracy: 0.5641 - val_loss: 0.1305 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2769 - accuracy: 0.5641 - val_loss: 0.1853 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2733 - accuracy: 0.5641 - val_loss: 0.1175 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2711 - accuracy: 0.5641 - val_loss: 0.1471 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2704 - accuracy: 0.5641 - val_loss: 0.1505 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2664 - accuracy: 0.5641 - val_loss: 0.1600 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 112ms/step - loss: 0.0934 - accuracy: 1.0000\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 5120, 32)         8832      \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 163840)            0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 163840)           655360    \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 163840)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               41943296  \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 256)              1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 256)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,608,512\n",
            "Trainable params: 42,280,320\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " model (Functional)          (None, 256)               42608512  \n",
            "                                                                 \n",
            " model_1 (Functional)        (None, 1)                 82305     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,690,817\n",
            "Trainable params: 42,362,625\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.5345 - accuracy: 0.4988 - val_loss: 0.3887 - val_accuracy: 0.8519\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.4062 - accuracy: 0.5475 - val_loss: 0.2450 - val_accuracy: 0.8519\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3818 - accuracy: 0.5544 - val_loss: 0.2143 - val_accuracy: 1.0000\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3425 - accuracy: 0.5764 - val_loss: 0.2149 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3286 - accuracy: 0.5787 - val_loss: 0.1947 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3239 - accuracy: 0.5752 - val_loss: 0.1977 - val_accuracy: 1.0000\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3140 - accuracy: 0.5822 - val_loss: 0.2000 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3211 - accuracy: 0.5810 - val_loss: 0.2471 - val_accuracy: 0.9259\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3063 - accuracy: 0.5845 - val_loss: 0.1577 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2903 - accuracy: 0.5833 - val_loss: 0.1933 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2999 - accuracy: 0.5856 - val_loss: 0.2074 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2909 - accuracy: 0.5856 - val_loss: 0.1633 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2896 - accuracy: 0.5822 - val_loss: 0.1445 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2904 - accuracy: 0.5833 - val_loss: 0.2004 - val_accuracy: 0.8889\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2870 - accuracy: 0.5856 - val_loss: 0.1501 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2804 - accuracy: 0.5845 - val_loss: 0.2128 - val_accuracy: 0.8889\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2781 - accuracy: 0.5845 - val_loss: 0.1170 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2655 - accuracy: 0.5845 - val_loss: 0.2123 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2782 - accuracy: 0.5856 - val_loss: 0.1773 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2671 - accuracy: 0.5856 - val_loss: 0.1721 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2713 - accuracy: 0.5845 - val_loss: 0.1688 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2717 - accuracy: 0.5856 - val_loss: 0.1118 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2697 - accuracy: 0.5856 - val_loss: 0.1349 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2662 - accuracy: 0.5856 - val_loss: 0.1468 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2773 - accuracy: 0.5833 - val_loss: 0.1934 - val_accuracy: 0.8889\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2644 - accuracy: 0.5856 - val_loss: 0.1211 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2691 - accuracy: 0.5845 - val_loss: 0.1162 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2667 - accuracy: 0.5856 - val_loss: 0.1305 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2650 - accuracy: 0.5856 - val_loss: 0.1947 - val_accuracy: 0.8889\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2629 - accuracy: 0.5845 - val_loss: 0.1813 - val_accuracy: 0.9259\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2637 - accuracy: 0.5856 - val_loss: 0.2392 - val_accuracy: 0.8519\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2690 - accuracy: 0.5845 - val_loss: 0.1529 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2650 - accuracy: 0.5845 - val_loss: 0.1589 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2592 - accuracy: 0.5856 - val_loss: 0.2135 - val_accuracy: 0.8889\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2622 - accuracy: 0.5856 - val_loss: 0.2071 - val_accuracy: 0.8519\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2537 - accuracy: 0.5845 - val_loss: 0.0917 - val_accuracy: 1.0000\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2544 - accuracy: 0.5845 - val_loss: 0.1889 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2562 - accuracy: 0.5856 - val_loss: 0.1541 - val_accuracy: 0.9259\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2505 - accuracy: 0.5856 - val_loss: 0.1276 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2525 - accuracy: 0.5856 - val_loss: 0.1133 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2478 - accuracy: 0.5856 - val_loss: 0.1383 - val_accuracy: 1.0000\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2523 - accuracy: 0.5856 - val_loss: 0.2079 - val_accuracy: 0.8889\n",
            "Epoch 43/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2540 - accuracy: 0.5856 - val_loss: 0.1447 - val_accuracy: 0.9630\n",
            "Epoch 44/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2619 - accuracy: 0.5856 - val_loss: 0.1307 - val_accuracy: 0.9630\n",
            "Epoch 45/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2519 - accuracy: 0.5856 - val_loss: 0.1612 - val_accuracy: 0.9630\n",
            "Epoch 46/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2508 - accuracy: 0.5856 - val_loss: 0.1989 - val_accuracy: 0.9259\n",
            "Epoch 47/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2549 - accuracy: 0.5856 - val_loss: 0.1283 - val_accuracy: 1.0000\n",
            "Epoch 48/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2533 - accuracy: 0.5856 - val_loss: 0.1266 - val_accuracy: 0.9630\n",
            "Epoch 49/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2484 - accuracy: 0.5856 - val_loss: 0.0875 - val_accuracy: 1.0000\n",
            "Epoch 50/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2514 - accuracy: 0.5856 - val_loss: 0.1116 - val_accuracy: 1.0000\n",
            "Epoch 51/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2489 - accuracy: 0.5856 - val_loss: 0.1184 - val_accuracy: 1.0000\n",
            "Epoch 52/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2506 - accuracy: 0.5856 - val_loss: 0.1194 - val_accuracy: 0.9630\n",
            "Epoch 53/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2490 - accuracy: 0.5856 - val_loss: 0.1055 - val_accuracy: 1.0000\n",
            "Epoch 54/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2509 - accuracy: 0.5856 - val_loss: 0.0818 - val_accuracy: 0.9630\n",
            "Epoch 55/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2503 - accuracy: 0.5856 - val_loss: 0.0789 - val_accuracy: 1.0000\n",
            "Epoch 56/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2491 - accuracy: 0.5856 - val_loss: 0.1101 - val_accuracy: 0.9630\n",
            "Epoch 57/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2448 - accuracy: 0.5856 - val_loss: 0.1096 - val_accuracy: 1.0000\n",
            "Epoch 58/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2479 - accuracy: 0.5856 - val_loss: 0.1044 - val_accuracy: 1.0000\n",
            "Epoch 59/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2483 - accuracy: 0.5856 - val_loss: 0.1573 - val_accuracy: 0.9259\n",
            "Epoch 60/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2487 - accuracy: 0.5856 - val_loss: 0.1322 - val_accuracy: 1.0000\n",
            "Epoch 61/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2484 - accuracy: 0.5856 - val_loss: 0.1111 - val_accuracy: 0.9630\n",
            "Epoch 62/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2470 - accuracy: 0.5856 - val_loss: 0.1065 - val_accuracy: 1.0000\n",
            "Epoch 63/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2472 - accuracy: 0.5856 - val_loss: 0.1056 - val_accuracy: 1.0000\n",
            "Epoch 64/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2460 - accuracy: 0.5856 - val_loss: 0.0885 - val_accuracy: 0.9630\n",
            "Epoch 65/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2451 - accuracy: 0.5856 - val_loss: 0.0879 - val_accuracy: 1.0000\n",
            "Epoch 66/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2492 - accuracy: 0.5856 - val_loss: 0.1123 - val_accuracy: 0.9630\n",
            "Epoch 67/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2428 - accuracy: 0.5856 - val_loss: 0.1135 - val_accuracy: 1.0000\n",
            "Epoch 68/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2426 - accuracy: 0.5856 - val_loss: 0.1140 - val_accuracy: 0.9630\n",
            "Epoch 69/100\n",
            "27/27 [==============================] - 8s 303ms/step - loss: 0.2440 - accuracy: 0.5856 - val_loss: 0.1255 - val_accuracy: 1.0000\n",
            "Epoch 70/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2395 - accuracy: 0.5856 - val_loss: 0.1227 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1797 - accuracy: 0.9706\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 5120, 32)         8832      \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 163840)            0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 163840)           655360    \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 163840)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               41943296  \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 256)              1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 256)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,608,512\n",
            "Trainable params: 42,280,320\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " model (Functional)          (None, 256)               42608512  \n",
            "                                                                 \n",
            " model_1 (Functional)        (None, 1)                 82305     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,690,817\n",
            "Trainable params: 42,362,625\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 330ms/step - loss: 0.5092 - accuracy: 0.5185 - val_loss: 0.1718 - val_accuracy: 0.9259\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3994 - accuracy: 0.5579 - val_loss: 0.1413 - val_accuracy: 1.0000\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3711 - accuracy: 0.5694 - val_loss: 0.1374 - val_accuracy: 1.0000\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3493 - accuracy: 0.5706 - val_loss: 0.2588 - val_accuracy: 0.8519\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3341 - accuracy: 0.5787 - val_loss: 0.1524 - val_accuracy: 1.0000\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3155 - accuracy: 0.5833 - val_loss: 0.1190 - val_accuracy: 1.0000\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2979 - accuracy: 0.5856 - val_loss: 0.1210 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3229 - accuracy: 0.5810 - val_loss: 0.1247 - val_accuracy: 1.0000\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2973 - accuracy: 0.5868 - val_loss: 0.1516 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2913 - accuracy: 0.5822 - val_loss: 0.1088 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2877 - accuracy: 0.5856 - val_loss: 0.1230 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2857 - accuracy: 0.5868 - val_loss: 0.0994 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2851 - accuracy: 0.5880 - val_loss: 0.0952 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2882 - accuracy: 0.5868 - val_loss: 0.0887 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2790 - accuracy: 0.5880 - val_loss: 0.0554 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2626 - accuracy: 0.5880 - val_loss: 0.0808 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2768 - accuracy: 0.5880 - val_loss: 0.0577 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2766 - accuracy: 0.5880 - val_loss: 0.1041 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2743 - accuracy: 0.5880 - val_loss: 0.1233 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2715 - accuracy: 0.5868 - val_loss: 0.0477 - val_accuracy: 1.0000\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2714 - accuracy: 0.5880 - val_loss: 0.0833 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2686 - accuracy: 0.5880 - val_loss: 0.1120 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2737 - accuracy: 0.5868 - val_loss: 0.0526 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2672 - accuracy: 0.5868 - val_loss: 0.0715 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2635 - accuracy: 0.5880 - val_loss: 0.0803 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2661 - accuracy: 0.5880 - val_loss: 0.0610 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2583 - accuracy: 0.5880 - val_loss: 0.0541 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2632 - accuracy: 0.5880 - val_loss: 0.0572 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2664 - accuracy: 0.5880 - val_loss: 0.0476 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2602 - accuracy: 0.5880 - val_loss: 0.0872 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2562 - accuracy: 0.5880 - val_loss: 0.0606 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2602 - accuracy: 0.5880 - val_loss: 0.0809 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2554 - accuracy: 0.5880 - val_loss: 0.0890 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2535 - accuracy: 0.5880 - val_loss: 0.0791 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2502 - accuracy: 0.5880 - val_loss: 0.0710 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2497 - accuracy: 0.5880 - val_loss: 0.1084 - val_accuracy: 1.0000\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2572 - accuracy: 0.5880 - val_loss: 0.0658 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2565 - accuracy: 0.5880 - val_loss: 0.0586 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2546 - accuracy: 0.5880 - val_loss: 0.0810 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2477 - accuracy: 0.5880 - val_loss: 0.0649 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2549 - accuracy: 0.5868 - val_loss: 0.0875 - val_accuracy: 1.0000\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2487 - accuracy: 0.5880 - val_loss: 0.0628 - val_accuracy: 1.0000\n",
            "Epoch 43/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2464 - accuracy: 0.5880 - val_loss: 0.0656 - val_accuracy: 1.0000\n",
            "Epoch 44/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2546 - accuracy: 0.5880 - val_loss: 0.0502 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 115ms/step - loss: 0.1584 - accuracy: 0.9412\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 5120, 32)         8832      \n",
            " l)                                                              \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 163840)            0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 163840)           655360    \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 163840)            0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               41943296  \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 256)              1024      \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " activation (Activation)     (None, 256)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,608,512\n",
            "Trainable params: 42,280,320\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 5120, 52)]        0         \n",
            "                                                                 \n",
            " model (Functional)          (None, 256)               42608512  \n",
            "                                                                 \n",
            " model_1 (Functional)        (None, 1)                 82305     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 42,690,817\n",
            "Trainable params: 42,362,625\n",
            "Non-trainable params: 328,192\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 332ms/step - loss: 0.5770 - accuracy: 0.5231 - val_loss: 0.3071 - val_accuracy: 0.8519\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.4127 - accuracy: 0.5741 - val_loss: 0.2264 - val_accuracy: 0.8519\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3520 - accuracy: 0.5903 - val_loss: 0.2103 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3256 - accuracy: 0.6123 - val_loss: 0.2012 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3142 - accuracy: 0.6146 - val_loss: 0.1667 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3049 - accuracy: 0.6123 - val_loss: 0.1485 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3100 - accuracy: 0.6169 - val_loss: 0.1915 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2903 - accuracy: 0.6169 - val_loss: 0.1256 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2995 - accuracy: 0.6169 - val_loss: 0.1479 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2823 - accuracy: 0.6181 - val_loss: 0.1330 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2760 - accuracy: 0.6181 - val_loss: 0.1455 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2725 - accuracy: 0.6204 - val_loss: 0.1138 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2750 - accuracy: 0.6192 - val_loss: 0.1458 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2677 - accuracy: 0.6204 - val_loss: 0.1483 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2747 - accuracy: 0.6204 - val_loss: 0.1158 - val_accuracy: 0.9259\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2672 - accuracy: 0.6204 - val_loss: 0.2057 - val_accuracy: 0.9259\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2732 - accuracy: 0.6192 - val_loss: 0.1123 - val_accuracy: 0.9259\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2627 - accuracy: 0.6204 - val_loss: 0.1254 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2733 - accuracy: 0.6192 - val_loss: 0.1299 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2635 - accuracy: 0.6192 - val_loss: 0.1320 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2544 - accuracy: 0.6204 - val_loss: 0.1303 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2600 - accuracy: 0.6204 - val_loss: 0.1328 - val_accuracy: 0.9259\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2629 - accuracy: 0.6204 - val_loss: 0.1489 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2502 - accuracy: 0.6204 - val_loss: 0.1436 - val_accuracy: 0.9259\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2563 - accuracy: 0.6204 - val_loss: 0.1602 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2537 - accuracy: 0.6204 - val_loss: 0.1402 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2537 - accuracy: 0.6204 - val_loss: 0.1101 - val_accuracy: 0.9259\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2493 - accuracy: 0.6192 - val_loss: 0.1287 - val_accuracy: 0.9259\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2463 - accuracy: 0.6204 - val_loss: 0.1349 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2475 - accuracy: 0.6204 - val_loss: 0.1430 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2464 - accuracy: 0.6204 - val_loss: 0.1776 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2454 - accuracy: 0.6204 - val_loss: 0.1637 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2485 - accuracy: 0.6204 - val_loss: 0.1194 - val_accuracy: 0.9259\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2452 - accuracy: 0.6204 - val_loss: 0.1440 - val_accuracy: 0.9259\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2392 - accuracy: 0.6204 - val_loss: 0.1883 - val_accuracy: 0.9259\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2508 - accuracy: 0.6204 - val_loss: 0.1503 - val_accuracy: 0.9259\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2446 - accuracy: 0.6204 - val_loss: 0.1158 - val_accuracy: 0.9259\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2365 - accuracy: 0.6204 - val_loss: 0.1599 - val_accuracy: 0.9259\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2375 - accuracy: 0.6204 - val_loss: 0.2531 - val_accuracy: 0.9259\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2450 - accuracy: 0.6204 - val_loss: 0.1420 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2374 - accuracy: 0.6204 - val_loss: 0.1335 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2368 - accuracy: 0.6204 - val_loss: 0.1513 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.0553 - accuracy: 1.0000\n",
            "WARNING:tensorflow:5 out of the last 9 calls to <function Model.make_predict_function.<locals>.predict_function at 0x7fee10091a70> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
            "[0.9823529362678528]\n",
            "[0.9789437689969607]\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.5331 - accuracy: 0.4941 - val_loss: 0.1570 - val_accuracy: 0.9630\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.4222 - accuracy: 0.5343 - val_loss: 0.2831 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3970 - accuracy: 0.5390 - val_loss: 0.1807 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3682 - accuracy: 0.5638 - val_loss: 0.1076 - val_accuracy: 1.0000\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3472 - accuracy: 0.5674 - val_loss: 0.1276 - val_accuracy: 1.0000\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3261 - accuracy: 0.5733 - val_loss: 0.1885 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3408 - accuracy: 0.5638 - val_loss: 0.0878 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3168 - accuracy: 0.5697 - val_loss: 0.0877 - val_accuracy: 1.0000\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3173 - accuracy: 0.5733 - val_loss: 0.0801 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.3074 - accuracy: 0.5745 - val_loss: 0.0868 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2984 - accuracy: 0.5757 - val_loss: 0.1122 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2969 - accuracy: 0.5733 - val_loss: 0.0959 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3009 - accuracy: 0.5768 - val_loss: 0.1366 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2987 - accuracy: 0.5768 - val_loss: 0.0642 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2895 - accuracy: 0.5768 - val_loss: 0.0698 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2931 - accuracy: 0.5768 - val_loss: 0.0532 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2924 - accuracy: 0.5757 - val_loss: 0.0624 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2837 - accuracy: 0.5768 - val_loss: 0.1132 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2736 - accuracy: 0.5768 - val_loss: 0.1142 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2832 - accuracy: 0.5745 - val_loss: 0.1183 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2850 - accuracy: 0.5757 - val_loss: 0.0623 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2833 - accuracy: 0.5768 - val_loss: 0.0786 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2842 - accuracy: 0.5757 - val_loss: 0.0482 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2759 - accuracy: 0.5768 - val_loss: 0.0871 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2788 - accuracy: 0.5768 - val_loss: 0.0773 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2764 - accuracy: 0.5768 - val_loss: 0.0959 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2727 - accuracy: 0.5768 - val_loss: 0.1041 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2772 - accuracy: 0.5768 - val_loss: 0.0671 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2759 - accuracy: 0.5768 - val_loss: 0.0770 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2656 - accuracy: 0.5768 - val_loss: 0.0859 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2671 - accuracy: 0.5768 - val_loss: 0.1037 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2664 - accuracy: 0.5768 - val_loss: 0.0973 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2727 - accuracy: 0.5768 - val_loss: 0.0965 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2631 - accuracy: 0.5768 - val_loss: 0.1033 - val_accuracy: 0.9630\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2662 - accuracy: 0.5768 - val_loss: 0.0886 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2651 - accuracy: 0.5768 - val_loss: 0.1135 - val_accuracy: 1.0000\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2705 - accuracy: 0.5757 - val_loss: 0.1114 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2618 - accuracy: 0.5768 - val_loss: 0.0949 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 111ms/step - loss: 0.0860 - accuracy: 0.9706\n",
            "WARNING:tensorflow:6 out of the last 11 calls to <function Model.make_predict_function.<locals>.predict_function at 0x7fee1013e200> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 332ms/step - loss: 0.4979 - accuracy: 0.5023 - val_loss: 0.4230 - val_accuracy: 0.7778\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.4035 - accuracy: 0.5359 - val_loss: 0.3897 - val_accuracy: 0.8148\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3882 - accuracy: 0.5451 - val_loss: 0.3042 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3556 - accuracy: 0.5521 - val_loss: 0.2845 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3449 - accuracy: 0.5567 - val_loss: 0.3078 - val_accuracy: 0.9259\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3326 - accuracy: 0.5544 - val_loss: 0.2694 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3173 - accuracy: 0.5590 - val_loss: 0.2586 - val_accuracy: 0.8889\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3154 - accuracy: 0.5590 - val_loss: 0.2056 - val_accuracy: 0.9259\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3163 - accuracy: 0.5590 - val_loss: 0.1710 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3036 - accuracy: 0.5625 - val_loss: 0.2196 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3051 - accuracy: 0.5625 - val_loss: 0.2033 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3047 - accuracy: 0.5613 - val_loss: 0.1889 - val_accuracy: 0.9259\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2983 - accuracy: 0.5625 - val_loss: 0.1923 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2975 - accuracy: 0.5613 - val_loss: 0.1679 - val_accuracy: 0.9259\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2995 - accuracy: 0.5613 - val_loss: 0.1721 - val_accuracy: 0.9259\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2932 - accuracy: 0.5625 - val_loss: 0.1594 - val_accuracy: 0.9259\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2902 - accuracy: 0.5625 - val_loss: 0.2170 - val_accuracy: 0.9259\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2930 - accuracy: 0.5625 - val_loss: 0.1354 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2818 - accuracy: 0.5625 - val_loss: 0.1647 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2838 - accuracy: 0.5625 - val_loss: 0.1596 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2896 - accuracy: 0.5613 - val_loss: 0.1565 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2879 - accuracy: 0.5625 - val_loss: 0.1654 - val_accuracy: 0.9259\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2812 - accuracy: 0.5625 - val_loss: 0.1751 - val_accuracy: 0.9259\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2747 - accuracy: 0.5625 - val_loss: 0.1772 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2804 - accuracy: 0.5625 - val_loss: 0.1294 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2767 - accuracy: 0.5625 - val_loss: 0.1665 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2733 - accuracy: 0.5625 - val_loss: 0.2260 - val_accuracy: 0.9259\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2798 - accuracy: 0.5625 - val_loss: 0.2068 - val_accuracy: 0.9259\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2819 - accuracy: 0.5625 - val_loss: 0.1866 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2782 - accuracy: 0.5625 - val_loss: 0.1677 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2730 - accuracy: 0.5625 - val_loss: 0.1771 - val_accuracy: 0.9259\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2693 - accuracy: 0.5625 - val_loss: 0.1564 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2684 - accuracy: 0.5625 - val_loss: 0.1537 - val_accuracy: 0.9259\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2736 - accuracy: 0.5625 - val_loss: 0.1516 - val_accuracy: 0.9259\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2773 - accuracy: 0.5625 - val_loss: 0.1498 - val_accuracy: 0.9259\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2718 - accuracy: 0.5625 - val_loss: 0.1888 - val_accuracy: 0.9259\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2638 - accuracy: 0.5625 - val_loss: 0.1580 - val_accuracy: 0.9259\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2683 - accuracy: 0.5625 - val_loss: 0.1280 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2682 - accuracy: 0.5625 - val_loss: 0.1322 - val_accuracy: 0.9630\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2637 - accuracy: 0.5625 - val_loss: 0.1191 - val_accuracy: 0.9259\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2662 - accuracy: 0.5625 - val_loss: 0.1529 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2657 - accuracy: 0.5625 - val_loss: 0.1370 - val_accuracy: 0.9630\n",
            "Epoch 43/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2635 - accuracy: 0.5625 - val_loss: 0.1453 - val_accuracy: 0.9630\n",
            "Epoch 44/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2643 - accuracy: 0.5625 - val_loss: 0.1983 - val_accuracy: 0.9259\n",
            "Epoch 45/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2671 - accuracy: 0.5625 - val_loss: 0.1912 - val_accuracy: 0.9630\n",
            "Epoch 46/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2604 - accuracy: 0.5625 - val_loss: 0.2392 - val_accuracy: 0.9259\n",
            "Epoch 47/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2656 - accuracy: 0.5625 - val_loss: 0.1713 - val_accuracy: 0.9630\n",
            "Epoch 48/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2648 - accuracy: 0.5625 - val_loss: 0.2610 - val_accuracy: 0.9259\n",
            "Epoch 49/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2626 - accuracy: 0.5625 - val_loss: 0.1717 - val_accuracy: 0.9259\n",
            "Epoch 50/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2607 - accuracy: 0.5625 - val_loss: 0.2553 - val_accuracy: 0.9259\n",
            "Epoch 51/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2592 - accuracy: 0.5625 - val_loss: 0.3473 - val_accuracy: 0.9259\n",
            "Epoch 52/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2608 - accuracy: 0.5625 - val_loss: 0.1948 - val_accuracy: 0.9259\n",
            "Epoch 53/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2614 - accuracy: 0.5625 - val_loss: 0.1964 - val_accuracy: 0.9259\n",
            "Epoch 54/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2636 - accuracy: 0.5625 - val_loss: 0.2213 - val_accuracy: 0.9259\n",
            "Epoch 55/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2647 - accuracy: 0.5625 - val_loss: 0.1794 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.0952 - accuracy: 1.0000\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 330ms/step - loss: 0.5522 - accuracy: 0.4921 - val_loss: 0.1066 - val_accuracy: 0.9630\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.4361 - accuracy: 0.5203 - val_loss: 0.2816 - val_accuracy: 0.7778\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3742 - accuracy: 0.5518 - val_loss: 0.2099 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3575 - accuracy: 0.5462 - val_loss: 0.1388 - val_accuracy: 1.0000\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3547 - accuracy: 0.5529 - val_loss: 0.1835 - val_accuracy: 0.9259\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3459 - accuracy: 0.5574 - val_loss: 0.2180 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3359 - accuracy: 0.5642 - val_loss: 0.1608 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3361 - accuracy: 0.5608 - val_loss: 0.2497 - val_accuracy: 0.8889\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3294 - accuracy: 0.5608 - val_loss: 0.1758 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 301ms/step - loss: 0.3101 - accuracy: 0.5653 - val_loss: 0.0968 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3120 - accuracy: 0.5664 - val_loss: 0.1062 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3001 - accuracy: 0.5676 - val_loss: 0.1114 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2999 - accuracy: 0.5653 - val_loss: 0.1430 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 301ms/step - loss: 0.3053 - accuracy: 0.5664 - val_loss: 0.0751 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3034 - accuracy: 0.5676 - val_loss: 0.1219 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2976 - accuracy: 0.5676 - val_loss: 0.0934 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3010 - accuracy: 0.5664 - val_loss: 0.0997 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2847 - accuracy: 0.5676 - val_loss: 0.0839 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2862 - accuracy: 0.5653 - val_loss: 0.0862 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2895 - accuracy: 0.5653 - val_loss: 0.0917 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2853 - accuracy: 0.5664 - val_loss: 0.1021 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 301ms/step - loss: 0.2897 - accuracy: 0.5676 - val_loss: 0.0678 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2796 - accuracy: 0.5676 - val_loss: 0.0923 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2846 - accuracy: 0.5676 - val_loss: 0.0936 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2795 - accuracy: 0.5676 - val_loss: 0.0958 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2773 - accuracy: 0.5676 - val_loss: 0.0709 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2798 - accuracy: 0.5676 - val_loss: 0.0898 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2803 - accuracy: 0.5676 - val_loss: 0.0599 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2698 - accuracy: 0.5676 - val_loss: 0.0763 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2685 - accuracy: 0.5676 - val_loss: 0.1120 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2709 - accuracy: 0.5676 - val_loss: 0.1482 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2794 - accuracy: 0.5676 - val_loss: 0.1208 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2708 - accuracy: 0.5676 - val_loss: 0.0867 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2755 - accuracy: 0.5676 - val_loss: 0.0761 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2675 - accuracy: 0.5676 - val_loss: 0.1522 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2658 - accuracy: 0.5676 - val_loss: 0.1204 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2675 - accuracy: 0.5676 - val_loss: 0.1042 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2744 - accuracy: 0.5676 - val_loss: 0.1405 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2625 - accuracy: 0.5676 - val_loss: 0.0769 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2656 - accuracy: 0.5676 - val_loss: 0.0900 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2698 - accuracy: 0.5676 - val_loss: 0.0991 - val_accuracy: 1.0000\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2642 - accuracy: 0.5676 - val_loss: 0.1060 - val_accuracy: 1.0000\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2615 - accuracy: 0.5676 - val_loss: 0.1243 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 114ms/step - loss: 0.0928 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 328ms/step - loss: 0.5280 - accuracy: 0.4863 - val_loss: 0.2693 - val_accuracy: 0.8889\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.4037 - accuracy: 0.5434 - val_loss: 0.2327 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3714 - accuracy: 0.5525 - val_loss: 0.2481 - val_accuracy: 0.9630\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3557 - accuracy: 0.5594 - val_loss: 0.2066 - val_accuracy: 0.9259\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3376 - accuracy: 0.5651 - val_loss: 0.2012 - val_accuracy: 0.9259\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3441 - accuracy: 0.5639 - val_loss: 0.2063 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3366 - accuracy: 0.5605 - val_loss: 0.1831 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3239 - accuracy: 0.5639 - val_loss: 0.1692 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3157 - accuracy: 0.5674 - val_loss: 0.1463 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3179 - accuracy: 0.5685 - val_loss: 0.2129 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3058 - accuracy: 0.5674 - val_loss: 0.1244 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3087 - accuracy: 0.5708 - val_loss: 0.1180 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2961 - accuracy: 0.5696 - val_loss: 0.0932 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2992 - accuracy: 0.5696 - val_loss: 0.1314 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2891 - accuracy: 0.5696 - val_loss: 0.0965 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2951 - accuracy: 0.5685 - val_loss: 0.1032 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3002 - accuracy: 0.5696 - val_loss: 0.0891 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2833 - accuracy: 0.5685 - val_loss: 0.0597 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2925 - accuracy: 0.5708 - val_loss: 0.0759 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2857 - accuracy: 0.5696 - val_loss: 0.0833 - val_accuracy: 1.0000\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2839 - accuracy: 0.5708 - val_loss: 0.1296 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2794 - accuracy: 0.5696 - val_loss: 0.0868 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2839 - accuracy: 0.5708 - val_loss: 0.0902 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2842 - accuracy: 0.5696 - val_loss: 0.0457 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2687 - accuracy: 0.5708 - val_loss: 0.0522 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2748 - accuracy: 0.5708 - val_loss: 0.0639 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2757 - accuracy: 0.5708 - val_loss: 0.0771 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2760 - accuracy: 0.5708 - val_loss: 0.0759 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2730 - accuracy: 0.5708 - val_loss: 0.0630 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2732 - accuracy: 0.5708 - val_loss: 0.0518 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2721 - accuracy: 0.5708 - val_loss: 0.0515 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2660 - accuracy: 0.5708 - val_loss: 0.0534 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2774 - accuracy: 0.5708 - val_loss: 0.0645 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2652 - accuracy: 0.5708 - val_loss: 0.0743 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2712 - accuracy: 0.5696 - val_loss: 0.0818 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2747 - accuracy: 0.5696 - val_loss: 0.0604 - val_accuracy: 1.0000\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2699 - accuracy: 0.5708 - val_loss: 0.0704 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2723 - accuracy: 0.5708 - val_loss: 0.0754 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2671 - accuracy: 0.5708 - val_loss: 0.0415 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2676 - accuracy: 0.5708 - val_loss: 0.0431 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2673 - accuracy: 0.5696 - val_loss: 0.0530 - val_accuracy: 1.0000\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2561 - accuracy: 0.5708 - val_loss: 0.0587 - val_accuracy: 1.0000\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2654 - accuracy: 0.5708 - val_loss: 0.1092 - val_accuracy: 1.0000\n",
            "Epoch 44/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2664 - accuracy: 0.5708 - val_loss: 0.0545 - val_accuracy: 1.0000\n",
            "Epoch 45/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2639 - accuracy: 0.5708 - val_loss: 0.0451 - val_accuracy: 1.0000\n",
            "Epoch 46/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2652 - accuracy: 0.5708 - val_loss: 0.0929 - val_accuracy: 1.0000\n",
            "Epoch 47/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2649 - accuracy: 0.5708 - val_loss: 0.0615 - val_accuracy: 1.0000\n",
            "Epoch 48/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2709 - accuracy: 0.5708 - val_loss: 0.0705 - val_accuracy: 1.0000\n",
            "Epoch 49/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2559 - accuracy: 0.5708 - val_loss: 0.0682 - val_accuracy: 1.0000\n",
            "Epoch 50/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2602 - accuracy: 0.5708 - val_loss: 0.0810 - val_accuracy: 1.0000\n",
            "Epoch 51/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2574 - accuracy: 0.5708 - val_loss: 0.0543 - val_accuracy: 1.0000\n",
            "Epoch 52/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2657 - accuracy: 0.5696 - val_loss: 0.0393 - val_accuracy: 1.0000\n",
            "Epoch 53/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2566 - accuracy: 0.5708 - val_loss: 0.0626 - val_accuracy: 1.0000\n",
            "Epoch 54/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2581 - accuracy: 0.5708 - val_loss: 0.0523 - val_accuracy: 1.0000\n",
            "Epoch 55/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2592 - accuracy: 0.5708 - val_loss: 0.0647 - val_accuracy: 1.0000\n",
            "Epoch 56/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2594 - accuracy: 0.5708 - val_loss: 0.0548 - val_accuracy: 1.0000\n",
            "Epoch 57/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2591 - accuracy: 0.5708 - val_loss: 0.1233 - val_accuracy: 0.9630\n",
            "Epoch 58/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2609 - accuracy: 0.5708 - val_loss: 0.0756 - val_accuracy: 1.0000\n",
            "Epoch 59/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2532 - accuracy: 0.5708 - val_loss: 0.0461 - val_accuracy: 1.0000\n",
            "Epoch 60/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2544 - accuracy: 0.5708 - val_loss: 0.0605 - val_accuracy: 1.0000\n",
            "Epoch 61/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2546 - accuracy: 0.5708 - val_loss: 0.0519 - val_accuracy: 1.0000\n",
            "Epoch 62/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2537 - accuracy: 0.5708 - val_loss: 0.1252 - val_accuracy: 1.0000\n",
            "Epoch 63/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2557 - accuracy: 0.5708 - val_loss: 0.1102 - val_accuracy: 1.0000\n",
            "Epoch 64/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2551 - accuracy: 0.5708 - val_loss: 0.1118 - val_accuracy: 1.0000\n",
            "Epoch 65/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2551 - accuracy: 0.5708 - val_loss: 0.0687 - val_accuracy: 1.0000\n",
            "Epoch 66/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2564 - accuracy: 0.5708 - val_loss: 0.1029 - val_accuracy: 1.0000\n",
            "Epoch 67/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2565 - accuracy: 0.5708 - val_loss: 0.0567 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 114ms/step - loss: 0.1641 - accuracy: 0.9412\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.5420 - accuracy: 0.5012 - val_loss: 0.4494 - val_accuracy: 0.8519\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3870 - accuracy: 0.5650 - val_loss: 0.1958 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3853 - accuracy: 0.5650 - val_loss: 0.2178 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3453 - accuracy: 0.5804 - val_loss: 0.2718 - val_accuracy: 0.8519\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3082 - accuracy: 0.5863 - val_loss: 0.2241 - val_accuracy: 0.8889\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3153 - accuracy: 0.5863 - val_loss: 0.2024 - val_accuracy: 0.9630\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3202 - accuracy: 0.5792 - val_loss: 0.1772 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2968 - accuracy: 0.5898 - val_loss: 0.1297 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2974 - accuracy: 0.5875 - val_loss: 0.1656 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2934 - accuracy: 0.5887 - val_loss: 0.1474 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2902 - accuracy: 0.5851 - val_loss: 0.1852 - val_accuracy: 0.8889\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2812 - accuracy: 0.5910 - val_loss: 0.1736 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2800 - accuracy: 0.5910 - val_loss: 0.1969 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2776 - accuracy: 0.5898 - val_loss: 0.1767 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2802 - accuracy: 0.5887 - val_loss: 0.1426 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2636 - accuracy: 0.5910 - val_loss: 0.1050 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2669 - accuracy: 0.5910 - val_loss: 0.0914 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2629 - accuracy: 0.5910 - val_loss: 0.1528 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2775 - accuracy: 0.5910 - val_loss: 0.1283 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2666 - accuracy: 0.5898 - val_loss: 0.1534 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2743 - accuracy: 0.5910 - val_loss: 0.0859 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2641 - accuracy: 0.5910 - val_loss: 0.0918 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2627 - accuracy: 0.5910 - val_loss: 0.1143 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2559 - accuracy: 0.5910 - val_loss: 0.1291 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2549 - accuracy: 0.5910 - val_loss: 0.0948 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2551 - accuracy: 0.5910 - val_loss: 0.1245 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2571 - accuracy: 0.5910 - val_loss: 0.1309 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2536 - accuracy: 0.5910 - val_loss: 0.0737 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2624 - accuracy: 0.5910 - val_loss: 0.0838 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2595 - accuracy: 0.5898 - val_loss: 0.1261 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2633 - accuracy: 0.5910 - val_loss: 0.1162 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2647 - accuracy: 0.5910 - val_loss: 0.1175 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2563 - accuracy: 0.5910 - val_loss: 0.1291 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2555 - accuracy: 0.5910 - val_loss: 0.1812 - val_accuracy: 0.9259\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2592 - accuracy: 0.5910 - val_loss: 0.1744 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2600 - accuracy: 0.5910 - val_loss: 0.1031 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2532 - accuracy: 0.5910 - val_loss: 0.1206 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2532 - accuracy: 0.5910 - val_loss: 0.1346 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2502 - accuracy: 0.5910 - val_loss: 0.0881 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2511 - accuracy: 0.5910 - val_loss: 0.1591 - val_accuracy: 0.9259\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.2541 - accuracy: 0.5910 - val_loss: 0.0984 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2474 - accuracy: 0.5910 - val_loss: 0.1135 - val_accuracy: 0.9630\n",
            "Epoch 43/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2524 - accuracy: 0.5910 - val_loss: 0.1020 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1101 - accuracy: 0.9706\n",
            "[0.9823529362678528, 0.970588219165802]\n",
            "[0.9789437689969607, 0.9654841563514089]\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 332ms/step - loss: 0.5387 - accuracy: 0.4976 - val_loss: 0.7518 - val_accuracy: 0.5185\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.4386 - accuracy: 0.5307 - val_loss: 0.1681 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3904 - accuracy: 0.5473 - val_loss: 0.3125 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3649 - accuracy: 0.5626 - val_loss: 0.1883 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3485 - accuracy: 0.5567 - val_loss: 0.1552 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3687 - accuracy: 0.5567 - val_loss: 0.1333 - val_accuracy: 1.0000\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3357 - accuracy: 0.5638 - val_loss: 0.1235 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3316 - accuracy: 0.5638 - val_loss: 0.1316 - val_accuracy: 1.0000\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3141 - accuracy: 0.5697 - val_loss: 0.0897 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3042 - accuracy: 0.5686 - val_loss: 0.0959 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2964 - accuracy: 0.5674 - val_loss: 0.1149 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3024 - accuracy: 0.5674 - val_loss: 0.0949 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3045 - accuracy: 0.5697 - val_loss: 0.0785 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2982 - accuracy: 0.5686 - val_loss: 0.1088 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2949 - accuracy: 0.5697 - val_loss: 0.1052 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3021 - accuracy: 0.5686 - val_loss: 0.0943 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2940 - accuracy: 0.5674 - val_loss: 0.0926 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2815 - accuracy: 0.5697 - val_loss: 0.1149 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2796 - accuracy: 0.5697 - val_loss: 0.0794 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2816 - accuracy: 0.5686 - val_loss: 0.0962 - val_accuracy: 1.0000\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2934 - accuracy: 0.5686 - val_loss: 0.1067 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2856 - accuracy: 0.5697 - val_loss: 0.0801 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2873 - accuracy: 0.5674 - val_loss: 0.0876 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2869 - accuracy: 0.5686 - val_loss: 0.0638 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2823 - accuracy: 0.5686 - val_loss: 0.0794 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2790 - accuracy: 0.5697 - val_loss: 0.0757 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2737 - accuracy: 0.5697 - val_loss: 0.1074 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2771 - accuracy: 0.5697 - val_loss: 0.0897 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2716 - accuracy: 0.5697 - val_loss: 0.0743 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2729 - accuracy: 0.5697 - val_loss: 0.1161 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2690 - accuracy: 0.5697 - val_loss: 0.0950 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2728 - accuracy: 0.5697 - val_loss: 0.0863 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2707 - accuracy: 0.5697 - val_loss: 0.1141 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2768 - accuracy: 0.5697 - val_loss: 0.0972 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2739 - accuracy: 0.5697 - val_loss: 0.1109 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2662 - accuracy: 0.5697 - val_loss: 0.1249 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2722 - accuracy: 0.5697 - val_loss: 0.1169 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2770 - accuracy: 0.5697 - val_loss: 0.1030 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2709 - accuracy: 0.5686 - val_loss: 0.1165 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.0634 - accuracy: 1.0000\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 328ms/step - loss: 0.5905 - accuracy: 0.4655 - val_loss: 0.2969 - val_accuracy: 0.9259\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.4155 - accuracy: 0.5276 - val_loss: 0.2556 - val_accuracy: 0.9259\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3864 - accuracy: 0.5402 - val_loss: 0.2450 - val_accuracy: 0.8519\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3704 - accuracy: 0.5529 - val_loss: 0.2672 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3411 - accuracy: 0.5598 - val_loss: 0.2205 - val_accuracy: 0.9259\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.3321 - accuracy: 0.5632 - val_loss: 0.2446 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.3316 - accuracy: 0.5609 - val_loss: 0.2333 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3281 - accuracy: 0.5598 - val_loss: 0.1849 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3254 - accuracy: 0.5598 - val_loss: 0.1809 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3105 - accuracy: 0.5655 - val_loss: 0.1614 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3064 - accuracy: 0.5644 - val_loss: 0.1632 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3012 - accuracy: 0.5667 - val_loss: 0.1589 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2977 - accuracy: 0.5644 - val_loss: 0.1864 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2960 - accuracy: 0.5644 - val_loss: 0.1575 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2969 - accuracy: 0.5667 - val_loss: 0.1709 - val_accuracy: 0.8889\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2943 - accuracy: 0.5667 - val_loss: 0.1536 - val_accuracy: 0.9259\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3008 - accuracy: 0.5667 - val_loss: 0.1589 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2891 - accuracy: 0.5678 - val_loss: 0.1580 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2941 - accuracy: 0.5667 - val_loss: 0.1499 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2847 - accuracy: 0.5678 - val_loss: 0.1621 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2803 - accuracy: 0.5678 - val_loss: 0.1721 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2880 - accuracy: 0.5667 - val_loss: 0.1164 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2748 - accuracy: 0.5678 - val_loss: 0.1466 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2852 - accuracy: 0.5667 - val_loss: 0.2053 - val_accuracy: 0.9259\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2785 - accuracy: 0.5667 - val_loss: 0.1677 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2714 - accuracy: 0.5678 - val_loss: 0.1738 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2676 - accuracy: 0.5678 - val_loss: 0.1122 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2789 - accuracy: 0.5678 - val_loss: 0.2146 - val_accuracy: 0.9259\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2720 - accuracy: 0.5678 - val_loss: 0.1836 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2699 - accuracy: 0.5678 - val_loss: 0.2033 - val_accuracy: 0.9259\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2733 - accuracy: 0.5678 - val_loss: 0.2432 - val_accuracy: 0.9259\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2697 - accuracy: 0.5678 - val_loss: 0.1748 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2691 - accuracy: 0.5678 - val_loss: 0.1846 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2927 - accuracy: 0.5678 - val_loss: 0.1424 - val_accuracy: 0.9630\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2809 - accuracy: 0.5667 - val_loss: 0.1580 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2809 - accuracy: 0.5667 - val_loss: 0.1666 - val_accuracy: 0.9259\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2745 - accuracy: 0.5655 - val_loss: 0.1733 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2715 - accuracy: 0.5667 - val_loss: 0.2560 - val_accuracy: 0.9259\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2698 - accuracy: 0.5678 - val_loss: 0.2632 - val_accuracy: 0.8889\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2669 - accuracy: 0.5678 - val_loss: 0.1414 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2618 - accuracy: 0.5678 - val_loss: 0.1578 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2623 - accuracy: 0.5678 - val_loss: 0.1578 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 114ms/step - loss: 0.1491 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 331ms/step - loss: 0.5147 - accuracy: 0.5138 - val_loss: 0.4097 - val_accuracy: 0.7037\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3816 - accuracy: 0.5690 - val_loss: 0.1213 - val_accuracy: 0.9630\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3678 - accuracy: 0.5678 - val_loss: 0.1689 - val_accuracy: 0.9630\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3574 - accuracy: 0.5759 - val_loss: 0.1425 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3411 - accuracy: 0.5862 - val_loss: 0.1735 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3157 - accuracy: 0.5828 - val_loss: 0.1291 - val_accuracy: 0.9630\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3176 - accuracy: 0.5851 - val_loss: 0.1165 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3219 - accuracy: 0.5862 - val_loss: 0.1219 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3165 - accuracy: 0.5874 - val_loss: 0.0772 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.3019 - accuracy: 0.5874 - val_loss: 0.0824 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2914 - accuracy: 0.5908 - val_loss: 0.0947 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2860 - accuracy: 0.5908 - val_loss: 0.0958 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2824 - accuracy: 0.5931 - val_loss: 0.0668 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2836 - accuracy: 0.5920 - val_loss: 0.0495 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2884 - accuracy: 0.5931 - val_loss: 0.0625 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2800 - accuracy: 0.5931 - val_loss: 0.0626 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2784 - accuracy: 0.5931 - val_loss: 0.0722 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2759 - accuracy: 0.5931 - val_loss: 0.0491 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2832 - accuracy: 0.5920 - val_loss: 0.0520 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2789 - accuracy: 0.5920 - val_loss: 0.0714 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2746 - accuracy: 0.5920 - val_loss: 0.1643 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2970 - accuracy: 0.5862 - val_loss: 0.0771 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2792 - accuracy: 0.5931 - val_loss: 0.0964 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2714 - accuracy: 0.5931 - val_loss: 0.0412 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2792 - accuracy: 0.5908 - val_loss: 0.0348 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2727 - accuracy: 0.5908 - val_loss: 0.0363 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2697 - accuracy: 0.5920 - val_loss: 0.0278 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2629 - accuracy: 0.5931 - val_loss: 0.0389 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2731 - accuracy: 0.5931 - val_loss: 0.0457 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2677 - accuracy: 0.5931 - val_loss: 0.0759 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2593 - accuracy: 0.5931 - val_loss: 0.0569 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2588 - accuracy: 0.5931 - val_loss: 0.0704 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2724 - accuracy: 0.5920 - val_loss: 0.0452 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2608 - accuracy: 0.5931 - val_loss: 0.0421 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2602 - accuracy: 0.5931 - val_loss: 0.0401 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2671 - accuracy: 0.5920 - val_loss: 0.0722 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2561 - accuracy: 0.5931 - val_loss: 0.0826 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2524 - accuracy: 0.5931 - val_loss: 0.0549 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2512 - accuracy: 0.5931 - val_loss: 0.1298 - val_accuracy: 0.9630\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2563 - accuracy: 0.5931 - val_loss: 0.0554 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2642 - accuracy: 0.5920 - val_loss: 0.0678 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2564 - accuracy: 0.5931 - val_loss: 0.1016 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 112ms/step - loss: 0.0750 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 335ms/step - loss: 0.5051 - accuracy: 0.5150 - val_loss: 0.2758 - val_accuracy: 0.8889\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3960 - accuracy: 0.5532 - val_loss: 0.3614 - val_accuracy: 0.7778\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3569 - accuracy: 0.5741 - val_loss: 0.2146 - val_accuracy: 0.9630\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3263 - accuracy: 0.5833 - val_loss: 0.2583 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3393 - accuracy: 0.5775 - val_loss: 0.1792 - val_accuracy: 1.0000\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3214 - accuracy: 0.5787 - val_loss: 0.2551 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3005 - accuracy: 0.5845 - val_loss: 0.1932 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3065 - accuracy: 0.5856 - val_loss: 0.1995 - val_accuracy: 1.0000\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2988 - accuracy: 0.5845 - val_loss: 0.2343 - val_accuracy: 0.8889\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2967 - accuracy: 0.5856 - val_loss: 0.1721 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2954 - accuracy: 0.5880 - val_loss: 0.2147 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2846 - accuracy: 0.5880 - val_loss: 0.1807 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2854 - accuracy: 0.5903 - val_loss: 0.2267 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2792 - accuracy: 0.5903 - val_loss: 0.1681 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2843 - accuracy: 0.5891 - val_loss: 0.2178 - val_accuracy: 0.8889\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2758 - accuracy: 0.5891 - val_loss: 0.1752 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2707 - accuracy: 0.5891 - val_loss: 0.1687 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2785 - accuracy: 0.5880 - val_loss: 0.2415 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2712 - accuracy: 0.5903 - val_loss: 0.2389 - val_accuracy: 0.8889\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2672 - accuracy: 0.5903 - val_loss: 0.1963 - val_accuracy: 0.8889\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2698 - accuracy: 0.5903 - val_loss: 0.2371 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2759 - accuracy: 0.5891 - val_loss: 0.3045 - val_accuracy: 0.8889\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2619 - accuracy: 0.5891 - val_loss: 0.1710 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2660 - accuracy: 0.5903 - val_loss: 0.1783 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2684 - accuracy: 0.5891 - val_loss: 0.3635 - val_accuracy: 0.8148\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2597 - accuracy: 0.5903 - val_loss: 0.1757 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2509 - accuracy: 0.5903 - val_loss: 0.1594 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2597 - accuracy: 0.5903 - val_loss: 0.2065 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2606 - accuracy: 0.5891 - val_loss: 0.2437 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2626 - accuracy: 0.5903 - val_loss: 0.2247 - val_accuracy: 0.9259\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2596 - accuracy: 0.5903 - val_loss: 0.1906 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2550 - accuracy: 0.5903 - val_loss: 0.1635 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 302ms/step - loss: 0.2576 - accuracy: 0.5903 - val_loss: 0.1205 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2573 - accuracy: 0.5903 - val_loss: 0.1625 - val_accuracy: 0.9630\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2531 - accuracy: 0.5903 - val_loss: 0.2273 - val_accuracy: 0.8889\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2528 - accuracy: 0.5903 - val_loss: 0.1543 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2521 - accuracy: 0.5903 - val_loss: 0.1508 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2520 - accuracy: 0.5903 - val_loss: 0.1525 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2515 - accuracy: 0.5903 - val_loss: 0.1320 - val_accuracy: 1.0000\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2516 - accuracy: 0.5903 - val_loss: 0.1558 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2602 - accuracy: 0.5903 - val_loss: 0.1676 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2481 - accuracy: 0.5903 - val_loss: 0.2122 - val_accuracy: 0.9259\n",
            "Epoch 43/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2473 - accuracy: 0.5903 - val_loss: 0.2235 - val_accuracy: 0.9630\n",
            "Epoch 44/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2472 - accuracy: 0.5903 - val_loss: 0.1841 - val_accuracy: 0.9630\n",
            "Epoch 45/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2451 - accuracy: 0.5903 - val_loss: 0.1949 - val_accuracy: 0.8889\n",
            "Epoch 46/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2487 - accuracy: 0.5903 - val_loss: 0.1991 - val_accuracy: 0.9259\n",
            "Epoch 47/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2534 - accuracy: 0.5903 - val_loss: 0.1477 - val_accuracy: 0.9630\n",
            "Epoch 48/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2492 - accuracy: 0.5903 - val_loss: 0.1590 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 112ms/step - loss: 0.1321 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 330ms/step - loss: 0.5447 - accuracy: 0.4851 - val_loss: 0.3478 - val_accuracy: 0.8148\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.4166 - accuracy: 0.5356 - val_loss: 0.4001 - val_accuracy: 0.7407\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3605 - accuracy: 0.5552 - val_loss: 0.3174 - val_accuracy: 0.8148\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3633 - accuracy: 0.5540 - val_loss: 0.2713 - val_accuracy: 0.8519\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3470 - accuracy: 0.5621 - val_loss: 0.2437 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3367 - accuracy: 0.5678 - val_loss: 0.2487 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3482 - accuracy: 0.5621 - val_loss: 0.2612 - val_accuracy: 0.8519\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3259 - accuracy: 0.5724 - val_loss: 0.1784 - val_accuracy: 1.0000\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3307 - accuracy: 0.5724 - val_loss: 0.2436 - val_accuracy: 0.8519\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3175 - accuracy: 0.5678 - val_loss: 0.2033 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3096 - accuracy: 0.5701 - val_loss: 0.2312 - val_accuracy: 0.8519\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3046 - accuracy: 0.5736 - val_loss: 0.2308 - val_accuracy: 0.8889\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3032 - accuracy: 0.5713 - val_loss: 0.1270 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3104 - accuracy: 0.5747 - val_loss: 0.1629 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2924 - accuracy: 0.5747 - val_loss: 0.1458 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2838 - accuracy: 0.5747 - val_loss: 0.1826 - val_accuracy: 0.9259\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2841 - accuracy: 0.5736 - val_loss: 0.1777 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2910 - accuracy: 0.5736 - val_loss: 0.3747 - val_accuracy: 0.7778\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3032 - accuracy: 0.5736 - val_loss: 0.1536 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2837 - accuracy: 0.5724 - val_loss: 0.1610 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2803 - accuracy: 0.5736 - val_loss: 0.2136 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2776 - accuracy: 0.5747 - val_loss: 0.1239 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2741 - accuracy: 0.5747 - val_loss: 0.1630 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2749 - accuracy: 0.5747 - val_loss: 0.1747 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2731 - accuracy: 0.5736 - val_loss: 0.2111 - val_accuracy: 0.8519\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2825 - accuracy: 0.5747 - val_loss: 0.1363 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2718 - accuracy: 0.5747 - val_loss: 0.1151 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2651 - accuracy: 0.5747 - val_loss: 0.1322 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2643 - accuracy: 0.5724 - val_loss: 0.1908 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2628 - accuracy: 0.5747 - val_loss: 0.1573 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2678 - accuracy: 0.5747 - val_loss: 0.1311 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2693 - accuracy: 0.5747 - val_loss: 0.1003 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2687 - accuracy: 0.5747 - val_loss: 0.1072 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2676 - accuracy: 0.5747 - val_loss: 0.0976 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2754 - accuracy: 0.5736 - val_loss: 0.1492 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2782 - accuracy: 0.5736 - val_loss: 0.1571 - val_accuracy: 0.9259\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2689 - accuracy: 0.5736 - val_loss: 0.1415 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2696 - accuracy: 0.5747 - val_loss: 0.1439 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2722 - accuracy: 0.5747 - val_loss: 0.1657 - val_accuracy: 0.9630\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2623 - accuracy: 0.5747 - val_loss: 0.1627 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2595 - accuracy: 0.5747 - val_loss: 0.1445 - val_accuracy: 0.9630\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 293ms/step - loss: 0.2685 - accuracy: 0.5747 - val_loss: 0.1604 - val_accuracy: 1.0000\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2619 - accuracy: 0.5747 - val_loss: 0.1407 - val_accuracy: 0.9630\n",
            "Epoch 44/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2548 - accuracy: 0.5747 - val_loss: 0.1637 - val_accuracy: 0.9630\n",
            "Epoch 45/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2543 - accuracy: 0.5747 - val_loss: 0.1358 - val_accuracy: 1.0000\n",
            "Epoch 46/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2590 - accuracy: 0.5747 - val_loss: 0.1443 - val_accuracy: 0.9630\n",
            "Epoch 47/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2591 - accuracy: 0.5747 - val_loss: 0.1143 - val_accuracy: 0.9630\n",
            "Epoch 48/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2548 - accuracy: 0.5747 - val_loss: 0.1809 - val_accuracy: 1.0000\n",
            "Epoch 49/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2619 - accuracy: 0.5747 - val_loss: 0.1726 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 114ms/step - loss: 0.0634 - accuracy: 1.0000\n",
            "[0.9823529362678528, 0.970588219165802, 0.9823529243469238]\n",
            "[0.9789437689969607, 0.9654841563514089, 0.9796508230180756]\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.5096 - accuracy: 0.5069 - val_loss: 0.3323 - val_accuracy: 0.7778\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3891 - accuracy: 0.5683 - val_loss: 0.1614 - val_accuracy: 0.9259\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3596 - accuracy: 0.5775 - val_loss: 0.2089 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3396 - accuracy: 0.5799 - val_loss: 0.2094 - val_accuracy: 1.0000\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3255 - accuracy: 0.5833 - val_loss: 0.1721 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3205 - accuracy: 0.5880 - val_loss: 0.1525 - val_accuracy: 1.0000\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3007 - accuracy: 0.5880 - val_loss: 0.1436 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3048 - accuracy: 0.5856 - val_loss: 0.1588 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2965 - accuracy: 0.5891 - val_loss: 0.0902 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3009 - accuracy: 0.5891 - val_loss: 0.1130 - val_accuracy: 1.0000\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2926 - accuracy: 0.5880 - val_loss: 0.0918 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2938 - accuracy: 0.5880 - val_loss: 0.0940 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2841 - accuracy: 0.5891 - val_loss: 0.0746 - val_accuracy: 1.0000\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2803 - accuracy: 0.5891 - val_loss: 0.1195 - val_accuracy: 1.0000\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2735 - accuracy: 0.5891 - val_loss: 0.0641 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2752 - accuracy: 0.5903 - val_loss: 0.0725 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2930 - accuracy: 0.5891 - val_loss: 0.0977 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2822 - accuracy: 0.5903 - val_loss: 0.0596 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2647 - accuracy: 0.5891 - val_loss: 0.1074 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2663 - accuracy: 0.5903 - val_loss: 0.0623 - val_accuracy: 1.0000\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2747 - accuracy: 0.5903 - val_loss: 0.0681 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2672 - accuracy: 0.5903 - val_loss: 0.0656 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2631 - accuracy: 0.5880 - val_loss: 0.0879 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2641 - accuracy: 0.5903 - val_loss: 0.0984 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2595 - accuracy: 0.5903 - val_loss: 0.0806 - val_accuracy: 1.0000\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2674 - accuracy: 0.5891 - val_loss: 0.0665 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2597 - accuracy: 0.5903 - val_loss: 0.0574 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2601 - accuracy: 0.5891 - val_loss: 0.1219 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2563 - accuracy: 0.5903 - val_loss: 0.1269 - val_accuracy: 1.0000\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2580 - accuracy: 0.5903 - val_loss: 0.1175 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2555 - accuracy: 0.5903 - val_loss: 0.0759 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2561 - accuracy: 0.5903 - val_loss: 0.1029 - val_accuracy: 1.0000\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2525 - accuracy: 0.5903 - val_loss: 0.1338 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2555 - accuracy: 0.5903 - val_loss: 0.0986 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2550 - accuracy: 0.5903 - val_loss: 0.1486 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2504 - accuracy: 0.5903 - val_loss: 0.1056 - val_accuracy: 1.0000\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2558 - accuracy: 0.5903 - val_loss: 0.0972 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2501 - accuracy: 0.5891 - val_loss: 0.1156 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2528 - accuracy: 0.5903 - val_loss: 0.1849 - val_accuracy: 0.9259\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2567 - accuracy: 0.5903 - val_loss: 0.1523 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2488 - accuracy: 0.5903 - val_loss: 0.1209 - val_accuracy: 1.0000\n",
            "Epoch 42/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2441 - accuracy: 0.5903 - val_loss: 0.1138 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1129 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 328ms/step - loss: 0.5471 - accuracy: 0.5023 - val_loss: 0.3356 - val_accuracy: 0.8519\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.3942 - accuracy: 0.5598 - val_loss: 0.2320 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3703 - accuracy: 0.5713 - val_loss: 0.2590 - val_accuracy: 0.8519\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3645 - accuracy: 0.5747 - val_loss: 0.3671 - val_accuracy: 0.8148\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3367 - accuracy: 0.5816 - val_loss: 0.2102 - val_accuracy: 0.9259\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3312 - accuracy: 0.5839 - val_loss: 0.2084 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3197 - accuracy: 0.5862 - val_loss: 0.1624 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3251 - accuracy: 0.5897 - val_loss: 0.1997 - val_accuracy: 0.8519\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3090 - accuracy: 0.5908 - val_loss: 0.1497 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2952 - accuracy: 0.5885 - val_loss: 0.1814 - val_accuracy: 0.8889\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2924 - accuracy: 0.5931 - val_loss: 0.2015 - val_accuracy: 0.8889\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.3050 - accuracy: 0.5920 - val_loss: 0.3112 - val_accuracy: 0.8519\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2927 - accuracy: 0.5920 - val_loss: 0.2333 - val_accuracy: 0.8889\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2870 - accuracy: 0.5920 - val_loss: 0.1028 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2859 - accuracy: 0.5931 - val_loss: 0.1300 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2931 - accuracy: 0.5897 - val_loss: 0.0953 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2885 - accuracy: 0.5908 - val_loss: 0.1292 - val_accuracy: 0.9259\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2752 - accuracy: 0.5954 - val_loss: 0.1263 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2770 - accuracy: 0.5954 - val_loss: 0.1057 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2780 - accuracy: 0.5931 - val_loss: 0.0834 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2881 - accuracy: 0.5908 - val_loss: 0.0731 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2669 - accuracy: 0.5943 - val_loss: 0.0856 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2727 - accuracy: 0.5943 - val_loss: 0.0861 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2645 - accuracy: 0.5954 - val_loss: 0.0878 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2675 - accuracy: 0.5954 - val_loss: 0.0830 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2699 - accuracy: 0.5954 - val_loss: 0.1171 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2685 - accuracy: 0.5943 - val_loss: 0.1159 - val_accuracy: 0.9259\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2630 - accuracy: 0.5931 - val_loss: 0.0954 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2653 - accuracy: 0.5954 - val_loss: 0.0867 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2606 - accuracy: 0.5931 - val_loss: 0.0812 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2751 - accuracy: 0.5931 - val_loss: 0.0656 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2607 - accuracy: 0.5943 - val_loss: 0.0967 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2642 - accuracy: 0.5931 - val_loss: 0.1069 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2561 - accuracy: 0.5954 - val_loss: 0.1223 - val_accuracy: 0.9630\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2551 - accuracy: 0.5954 - val_loss: 0.1401 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2569 - accuracy: 0.5943 - val_loss: 0.1923 - val_accuracy: 0.9259\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2719 - accuracy: 0.5954 - val_loss: 0.1149 - val_accuracy: 0.9259\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2596 - accuracy: 0.5954 - val_loss: 0.1185 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2565 - accuracy: 0.5954 - val_loss: 0.0902 - val_accuracy: 0.9630\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2487 - accuracy: 0.5954 - val_loss: 0.1004 - val_accuracy: 0.9630\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2520 - accuracy: 0.5954 - val_loss: 0.1055 - val_accuracy: 0.9259\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2571 - accuracy: 0.5954 - val_loss: 0.1140 - val_accuracy: 0.9630\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2537 - accuracy: 0.5954 - val_loss: 0.1136 - val_accuracy: 0.9630\n",
            "Epoch 44/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2564 - accuracy: 0.5954 - val_loss: 0.1384 - val_accuracy: 0.9630\n",
            "Epoch 45/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2506 - accuracy: 0.5943 - val_loss: 0.1426 - val_accuracy: 0.9259\n",
            "Epoch 46/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2652 - accuracy: 0.5931 - val_loss: 0.0899 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 114ms/step - loss: 0.0514 - accuracy: 1.0000\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 330ms/step - loss: 0.5354 - accuracy: 0.4732 - val_loss: 0.5607 - val_accuracy: 0.7407\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3867 - accuracy: 0.5466 - val_loss: 0.2261 - val_accuracy: 0.9630\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3758 - accuracy: 0.5455 - val_loss: 0.3623 - val_accuracy: 0.7778\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3579 - accuracy: 0.5559 - val_loss: 0.2981 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3465 - accuracy: 0.5641 - val_loss: 0.2269 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3322 - accuracy: 0.5606 - val_loss: 0.2064 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3165 - accuracy: 0.5664 - val_loss: 0.1939 - val_accuracy: 0.9259\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3232 - accuracy: 0.5653 - val_loss: 0.1691 - val_accuracy: 0.9259\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3104 - accuracy: 0.5664 - val_loss: 0.2198 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3135 - accuracy: 0.5664 - val_loss: 0.1470 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3034 - accuracy: 0.5653 - val_loss: 0.1388 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2905 - accuracy: 0.5688 - val_loss: 0.2031 - val_accuracy: 0.9259\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2968 - accuracy: 0.5676 - val_loss: 0.1543 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2987 - accuracy: 0.5676 - val_loss: 0.1520 - val_accuracy: 0.9259\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2917 - accuracy: 0.5688 - val_loss: 0.2169 - val_accuracy: 0.9259\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2832 - accuracy: 0.5676 - val_loss: 0.1315 - val_accuracy: 0.9259\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2834 - accuracy: 0.5676 - val_loss: 0.1696 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2833 - accuracy: 0.5688 - val_loss: 0.1306 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2953 - accuracy: 0.5676 - val_loss: 0.1180 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2821 - accuracy: 0.5688 - val_loss: 0.1068 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2800 - accuracy: 0.5688 - val_loss: 0.0955 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2796 - accuracy: 0.5688 - val_loss: 0.0980 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2779 - accuracy: 0.5688 - val_loss: 0.1137 - val_accuracy: 1.0000\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2776 - accuracy: 0.5688 - val_loss: 0.1005 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2715 - accuracy: 0.5688 - val_loss: 0.1191 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2676 - accuracy: 0.5688 - val_loss: 0.1454 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2746 - accuracy: 0.5688 - val_loss: 0.1472 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2678 - accuracy: 0.5688 - val_loss: 0.1541 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2703 - accuracy: 0.5688 - val_loss: 0.1788 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2712 - accuracy: 0.5676 - val_loss: 0.1771 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2678 - accuracy: 0.5676 - val_loss: 0.2450 - val_accuracy: 0.8519\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2705 - accuracy: 0.5688 - val_loss: 0.1700 - val_accuracy: 0.9259\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2656 - accuracy: 0.5688 - val_loss: 0.1646 - val_accuracy: 0.9259\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2675 - accuracy: 0.5688 - val_loss: 0.1645 - val_accuracy: 0.9259\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2681 - accuracy: 0.5688 - val_loss: 0.1481 - val_accuracy: 0.9259\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2601 - accuracy: 0.5688 - val_loss: 0.2083 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1127 - accuracy: 1.0000\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 329ms/step - loss: 0.6007 - accuracy: 0.4806 - val_loss: 0.6450 - val_accuracy: 0.7037\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.4322 - accuracy: 0.5365 - val_loss: 0.3426 - val_accuracy: 0.8519\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3696 - accuracy: 0.5685 - val_loss: 0.3036 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 301ms/step - loss: 0.3458 - accuracy: 0.5776 - val_loss: 0.2842 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.3375 - accuracy: 0.5856 - val_loss: 0.2748 - val_accuracy: 0.8889\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3248 - accuracy: 0.5833 - val_loss: 0.2774 - val_accuracy: 0.8519\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3199 - accuracy: 0.5856 - val_loss: 0.2547 - val_accuracy: 0.8889\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3114 - accuracy: 0.5822 - val_loss: 0.2139 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2979 - accuracy: 0.5936 - val_loss: 0.2438 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2966 - accuracy: 0.5890 - val_loss: 0.1742 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2852 - accuracy: 0.5913 - val_loss: 0.1688 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2968 - accuracy: 0.5936 - val_loss: 0.1726 - val_accuracy: 0.9259\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2807 - accuracy: 0.5902 - val_loss: 0.2447 - val_accuracy: 0.8148\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2824 - accuracy: 0.5936 - val_loss: 0.1790 - val_accuracy: 0.8889\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2761 - accuracy: 0.5936 - val_loss: 0.0924 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 294ms/step - loss: 0.2714 - accuracy: 0.5936 - val_loss: 0.1430 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2696 - accuracy: 0.5936 - val_loss: 0.1433 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.2700 - accuracy: 0.5913 - val_loss: 0.0735 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2701 - accuracy: 0.5936 - val_loss: 0.2787 - val_accuracy: 0.8148\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2686 - accuracy: 0.5925 - val_loss: 0.1191 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2638 - accuracy: 0.5936 - val_loss: 0.0940 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2667 - accuracy: 0.5936 - val_loss: 0.1089 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2726 - accuracy: 0.5913 - val_loss: 0.1100 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2639 - accuracy: 0.5936 - val_loss: 0.1260 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2714 - accuracy: 0.5925 - val_loss: 0.1227 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2743 - accuracy: 0.5936 - val_loss: 0.0925 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2604 - accuracy: 0.5936 - val_loss: 0.0927 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2549 - accuracy: 0.5936 - val_loss: 0.1060 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2583 - accuracy: 0.5936 - val_loss: 0.1832 - val_accuracy: 0.8889\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2573 - accuracy: 0.5936 - val_loss: 0.2295 - val_accuracy: 0.8889\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2578 - accuracy: 0.5936 - val_loss: 0.1101 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2544 - accuracy: 0.5936 - val_loss: 0.1035 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2553 - accuracy: 0.5936 - val_loss: 0.1183 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 111ms/step - loss: 0.1125 - accuracy: 0.9412\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.4801 - accuracy: 0.5139 - val_loss: 0.1972 - val_accuracy: 0.8889\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3855 - accuracy: 0.5613 - val_loss: 0.2423 - val_accuracy: 0.8519\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3681 - accuracy: 0.5671 - val_loss: 0.2005 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3490 - accuracy: 0.5775 - val_loss: 0.2332 - val_accuracy: 0.8889\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3272 - accuracy: 0.5764 - val_loss: 0.1768 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3189 - accuracy: 0.5810 - val_loss: 0.2326 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3212 - accuracy: 0.5787 - val_loss: 0.2092 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3030 - accuracy: 0.5845 - val_loss: 0.1425 - val_accuracy: 0.9259\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2887 - accuracy: 0.5868 - val_loss: 0.1282 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2939 - accuracy: 0.5880 - val_loss: 0.1490 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2886 - accuracy: 0.5856 - val_loss: 0.1153 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2890 - accuracy: 0.5868 - val_loss: 0.1271 - val_accuracy: 0.9630\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2848 - accuracy: 0.5868 - val_loss: 0.0897 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2798 - accuracy: 0.5880 - val_loss: 0.0920 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2758 - accuracy: 0.5868 - val_loss: 0.1397 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2811 - accuracy: 0.5880 - val_loss: 0.0863 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2846 - accuracy: 0.5856 - val_loss: 0.1096 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2756 - accuracy: 0.5880 - val_loss: 0.1325 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2717 - accuracy: 0.5880 - val_loss: 0.1618 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2748 - accuracy: 0.5880 - val_loss: 0.0808 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2834 - accuracy: 0.5868 - val_loss: 0.1280 - val_accuracy: 0.9630\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2769 - accuracy: 0.5880 - val_loss: 0.0790 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2717 - accuracy: 0.5868 - val_loss: 0.1146 - val_accuracy: 0.9259\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2680 - accuracy: 0.5880 - val_loss: 0.0951 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2645 - accuracy: 0.5880 - val_loss: 0.0817 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2654 - accuracy: 0.5880 - val_loss: 0.0749 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2738 - accuracy: 0.5880 - val_loss: 0.1238 - val_accuracy: 0.9630\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2617 - accuracy: 0.5880 - val_loss: 0.1176 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2615 - accuracy: 0.5868 - val_loss: 0.0981 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2595 - accuracy: 0.5868 - val_loss: 0.0932 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2605 - accuracy: 0.5880 - val_loss: 0.0940 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2622 - accuracy: 0.5880 - val_loss: 0.1210 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2553 - accuracy: 0.5880 - val_loss: 0.0933 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2640 - accuracy: 0.5880 - val_loss: 0.0759 - val_accuracy: 0.9630\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2643 - accuracy: 0.5880 - val_loss: 0.1091 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2501 - accuracy: 0.5880 - val_loss: 0.0885 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2642 - accuracy: 0.5880 - val_loss: 0.1060 - val_accuracy: 0.9630\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2576 - accuracy: 0.5880 - val_loss: 0.0984 - val_accuracy: 0.9630\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2571 - accuracy: 0.5880 - val_loss: 0.1354 - val_accuracy: 0.9259\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2553 - accuracy: 0.5880 - val_loss: 0.1345 - val_accuracy: 0.9259\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2518 - accuracy: 0.5880 - val_loss: 0.1236 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 109ms/step - loss: 0.1531 - accuracy: 0.9706\n",
            "[0.9823529362678528, 0.970588219165802, 0.9823529243469238, 0.9764705777168274]\n",
            "[0.9789437689969607, 0.9654841563514089, 0.9796508230180756, 0.9736615994756604]\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 330ms/step - loss: 0.5361 - accuracy: 0.5390 - val_loss: 0.7576 - val_accuracy: 0.7407\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3755 - accuracy: 0.5863 - val_loss: 0.3101 - val_accuracy: 0.9259\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3471 - accuracy: 0.6028 - val_loss: 0.2927 - val_accuracy: 0.9259\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3364 - accuracy: 0.6064 - val_loss: 0.3139 - val_accuracy: 0.9259\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.3362 - accuracy: 0.6017 - val_loss: 0.3108 - val_accuracy: 0.8889\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3024 - accuracy: 0.6099 - val_loss: 0.2491 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3010 - accuracy: 0.6147 - val_loss: 0.4148 - val_accuracy: 0.8148\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 294ms/step - loss: 0.3063 - accuracy: 0.6052 - val_loss: 0.2494 - val_accuracy: 0.9259\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2847 - accuracy: 0.6158 - val_loss: 0.2190 - val_accuracy: 0.9259\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.2757 - accuracy: 0.6170 - val_loss: 0.2101 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2736 - accuracy: 0.6147 - val_loss: 0.2051 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2600 - accuracy: 0.6158 - val_loss: 0.2203 - val_accuracy: 0.9259\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2747 - accuracy: 0.6158 - val_loss: 0.2589 - val_accuracy: 0.8889\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2601 - accuracy: 0.6170 - val_loss: 0.1896 - val_accuracy: 0.9259\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2640 - accuracy: 0.6170 - val_loss: 0.2840 - val_accuracy: 0.8889\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2683 - accuracy: 0.6158 - val_loss: 0.1138 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2599 - accuracy: 0.6170 - val_loss: 0.2903 - val_accuracy: 0.9259\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2532 - accuracy: 0.6158 - val_loss: 0.2172 - val_accuracy: 0.9259\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2540 - accuracy: 0.6170 - val_loss: 0.1831 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2481 - accuracy: 0.6170 - val_loss: 0.1259 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2516 - accuracy: 0.6170 - val_loss: 0.3034 - val_accuracy: 0.8519\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2500 - accuracy: 0.6170 - val_loss: 0.1569 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2483 - accuracy: 0.6170 - val_loss: 0.1502 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2476 - accuracy: 0.6170 - val_loss: 0.1620 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2473 - accuracy: 0.6170 - val_loss: 0.1928 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2477 - accuracy: 0.6170 - val_loss: 0.1612 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2471 - accuracy: 0.6170 - val_loss: 0.2686 - val_accuracy: 0.9259\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2403 - accuracy: 0.6170 - val_loss: 0.1227 - val_accuracy: 0.9630\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2491 - accuracy: 0.6158 - val_loss: 0.1641 - val_accuracy: 0.9259\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2439 - accuracy: 0.6170 - val_loss: 0.2041 - val_accuracy: 0.9259\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2431 - accuracy: 0.6170 - val_loss: 0.1723 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1073 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 330ms/step - loss: 0.5295 - accuracy: 0.4873 - val_loss: 0.2823 - val_accuracy: 0.8148\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.4143 - accuracy: 0.5312 - val_loss: 0.1787 - val_accuracy: 0.9259\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3859 - accuracy: 0.5382 - val_loss: 0.2192 - val_accuracy: 0.9630\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3507 - accuracy: 0.5521 - val_loss: 0.1852 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3455 - accuracy: 0.5532 - val_loss: 0.1827 - val_accuracy: 1.0000\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3460 - accuracy: 0.5567 - val_loss: 0.2545 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3229 - accuracy: 0.5590 - val_loss: 0.1804 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3250 - accuracy: 0.5544 - val_loss: 0.2387 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3389 - accuracy: 0.5544 - val_loss: 0.1654 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 299ms/step - loss: 0.3130 - accuracy: 0.5602 - val_loss: 0.1606 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.3103 - accuracy: 0.5625 - val_loss: 0.1483 - val_accuracy: 0.9259\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2985 - accuracy: 0.5625 - val_loss: 0.1237 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2939 - accuracy: 0.5625 - val_loss: 0.2064 - val_accuracy: 0.9259\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2854 - accuracy: 0.5613 - val_loss: 0.1545 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2927 - accuracy: 0.5625 - val_loss: 0.4076 - val_accuracy: 0.8889\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2977 - accuracy: 0.5613 - val_loss: 0.1491 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2951 - accuracy: 0.5613 - val_loss: 0.1351 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2873 - accuracy: 0.5602 - val_loss: 0.1360 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2883 - accuracy: 0.5602 - val_loss: 0.2736 - val_accuracy: 0.9259\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 295ms/step - loss: 0.2888 - accuracy: 0.5625 - val_loss: 0.1580 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2872 - accuracy: 0.5625 - val_loss: 0.2377 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2851 - accuracy: 0.5625 - val_loss: 0.1382 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2914 - accuracy: 0.5613 - val_loss: 0.1808 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2818 - accuracy: 0.5613 - val_loss: 0.1375 - val_accuracy: 0.9259\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2759 - accuracy: 0.5625 - val_loss: 0.2011 - val_accuracy: 0.9259\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2758 - accuracy: 0.5625 - val_loss: 0.2176 - val_accuracy: 0.9259\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2735 - accuracy: 0.5613 - val_loss: 0.2184 - val_accuracy: 0.9630\n",
            "2/2 [==============================] - 1s 111ms/step - loss: 0.1257 - accuracy: 1.0000\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.4930 - accuracy: 0.5140 - val_loss: 0.3833 - val_accuracy: 0.8148\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.4009 - accuracy: 0.5524 - val_loss: 0.2669 - val_accuracy: 0.8148\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3699 - accuracy: 0.5455 - val_loss: 0.3180 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3438 - accuracy: 0.5571 - val_loss: 0.2914 - val_accuracy: 0.8148\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3352 - accuracy: 0.5629 - val_loss: 0.3148 - val_accuracy: 0.8889\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3276 - accuracy: 0.5664 - val_loss: 0.2589 - val_accuracy: 0.9259\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3135 - accuracy: 0.5688 - val_loss: 0.2724 - val_accuracy: 0.8889\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3097 - accuracy: 0.5664 - val_loss: 0.2435 - val_accuracy: 0.8889\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 302ms/step - loss: 0.3044 - accuracy: 0.5688 - val_loss: 0.2004 - val_accuracy: 0.9630\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3038 - accuracy: 0.5676 - val_loss: 0.2060 - val_accuracy: 0.9259\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2977 - accuracy: 0.5699 - val_loss: 0.2670 - val_accuracy: 0.8889\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2931 - accuracy: 0.5711 - val_loss: 0.2345 - val_accuracy: 0.8519\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3049 - accuracy: 0.5711 - val_loss: 0.2822 - val_accuracy: 0.8889\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2851 - accuracy: 0.5711 - val_loss: 0.2003 - val_accuracy: 0.8889\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2825 - accuracy: 0.5711 - val_loss: 0.1812 - val_accuracy: 0.9259\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2792 - accuracy: 0.5711 - val_loss: 0.2941 - val_accuracy: 0.8519\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2799 - accuracy: 0.5711 - val_loss: 0.3047 - val_accuracy: 0.8889\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2896 - accuracy: 0.5711 - val_loss: 0.1977 - val_accuracy: 0.8889\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2848 - accuracy: 0.5711 - val_loss: 0.2049 - val_accuracy: 0.8889\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 303ms/step - loss: 0.2850 - accuracy: 0.5711 - val_loss: 0.1627 - val_accuracy: 0.9259\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2757 - accuracy: 0.5711 - val_loss: 0.2212 - val_accuracy: 0.8148\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2745 - accuracy: 0.5711 - val_loss: 0.3574 - val_accuracy: 0.8889\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2662 - accuracy: 0.5711 - val_loss: 0.2825 - val_accuracy: 0.8889\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2673 - accuracy: 0.5711 - val_loss: 0.1664 - val_accuracy: 0.9259\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2651 - accuracy: 0.5711 - val_loss: 0.3940 - val_accuracy: 0.8889\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2681 - accuracy: 0.5711 - val_loss: 0.2924 - val_accuracy: 0.8889\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2708 - accuracy: 0.5699 - val_loss: 0.2396 - val_accuracy: 0.8889\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2718 - accuracy: 0.5699 - val_loss: 0.2157 - val_accuracy: 0.9259\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2691 - accuracy: 0.5711 - val_loss: 0.2611 - val_accuracy: 0.8889\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2668 - accuracy: 0.5711 - val_loss: 0.2466 - val_accuracy: 0.8889\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2694 - accuracy: 0.5711 - val_loss: 0.2920 - val_accuracy: 0.8889\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2621 - accuracy: 0.5711 - val_loss: 0.3288 - val_accuracy: 0.8889\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2613 - accuracy: 0.5711 - val_loss: 0.3697 - val_accuracy: 0.8889\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2696 - accuracy: 0.5711 - val_loss: 0.2752 - val_accuracy: 0.8889\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2628 - accuracy: 0.5711 - val_loss: 0.2633 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 113ms/step - loss: 0.1262 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "27/27 [==============================] - 12s 331ms/step - loss: 0.5346 - accuracy: 0.4872 - val_loss: 0.2258 - val_accuracy: 0.9259\n",
            "Epoch 2/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.4084 - accuracy: 0.5501 - val_loss: 0.2694 - val_accuracy: 0.8519\n",
            "Epoch 3/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.3708 - accuracy: 0.5723 - val_loss: 0.1992 - val_accuracy: 0.9630\n",
            "Epoch 4/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.3520 - accuracy: 0.5746 - val_loss: 0.2060 - val_accuracy: 0.9630\n",
            "Epoch 5/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.3281 - accuracy: 0.5851 - val_loss: 0.3434 - val_accuracy: 0.7778\n",
            "Epoch 6/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3359 - accuracy: 0.5816 - val_loss: 0.2327 - val_accuracy: 0.8889\n",
            "Epoch 7/100\n",
            "27/27 [==============================] - 8s 302ms/step - loss: 0.3211 - accuracy: 0.5816 - val_loss: 0.1691 - val_accuracy: 1.0000\n",
            "Epoch 8/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.3103 - accuracy: 0.5851 - val_loss: 0.2391 - val_accuracy: 0.8889\n",
            "Epoch 9/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2973 - accuracy: 0.5839 - val_loss: 0.1722 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2975 - accuracy: 0.5862 - val_loss: 0.1740 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2967 - accuracy: 0.5874 - val_loss: 0.1309 - val_accuracy: 1.0000\n",
            "Epoch 12/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2932 - accuracy: 0.5839 - val_loss: 0.1633 - val_accuracy: 1.0000\n",
            "Epoch 13/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2985 - accuracy: 0.5851 - val_loss: 0.1384 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2856 - accuracy: 0.5862 - val_loss: 0.1249 - val_accuracy: 0.9630\n",
            "Epoch 15/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2863 - accuracy: 0.5874 - val_loss: 0.1197 - val_accuracy: 1.0000\n",
            "Epoch 16/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2929 - accuracy: 0.5851 - val_loss: 0.1274 - val_accuracy: 1.0000\n",
            "Epoch 17/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2730 - accuracy: 0.5874 - val_loss: 0.0903 - val_accuracy: 1.0000\n",
            "Epoch 18/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2767 - accuracy: 0.5874 - val_loss: 0.0896 - val_accuracy: 1.0000\n",
            "Epoch 19/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2670 - accuracy: 0.5874 - val_loss: 0.0919 - val_accuracy: 1.0000\n",
            "Epoch 20/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2639 - accuracy: 0.5874 - val_loss: 0.0885 - val_accuracy: 1.0000\n",
            "Epoch 21/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2708 - accuracy: 0.5874 - val_loss: 0.0866 - val_accuracy: 1.0000\n",
            "Epoch 22/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2732 - accuracy: 0.5862 - val_loss: 0.0949 - val_accuracy: 1.0000\n",
            "Epoch 23/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2691 - accuracy: 0.5851 - val_loss: 0.1022 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "27/27 [==============================] - 8s 300ms/step - loss: 0.2647 - accuracy: 0.5874 - val_loss: 0.0731 - val_accuracy: 1.0000\n",
            "Epoch 25/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2659 - accuracy: 0.5874 - val_loss: 0.0897 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "27/27 [==============================] - 8s 301ms/step - loss: 0.2637 - accuracy: 0.5874 - val_loss: 0.0714 - val_accuracy: 1.0000\n",
            "Epoch 27/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2629 - accuracy: 0.5874 - val_loss: 0.1142 - val_accuracy: 1.0000\n",
            "Epoch 28/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2611 - accuracy: 0.5874 - val_loss: 0.1102 - val_accuracy: 1.0000\n",
            "Epoch 29/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2603 - accuracy: 0.5862 - val_loss: 0.1448 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2568 - accuracy: 0.5874 - val_loss: 0.0991 - val_accuracy: 1.0000\n",
            "Epoch 31/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2678 - accuracy: 0.5874 - val_loss: 0.0981 - val_accuracy: 1.0000\n",
            "Epoch 32/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2704 - accuracy: 0.5874 - val_loss: 0.1675 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2575 - accuracy: 0.5874 - val_loss: 0.1015 - val_accuracy: 1.0000\n",
            "Epoch 34/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2617 - accuracy: 0.5874 - val_loss: 0.1189 - val_accuracy: 1.0000\n",
            "Epoch 35/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2571 - accuracy: 0.5874 - val_loss: 0.0994 - val_accuracy: 1.0000\n",
            "Epoch 36/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2624 - accuracy: 0.5862 - val_loss: 0.1630 - val_accuracy: 0.9630\n",
            "Epoch 37/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2583 - accuracy: 0.5874 - val_loss: 0.1056 - val_accuracy: 1.0000\n",
            "Epoch 38/100\n",
            "27/27 [==============================] - 8s 297ms/step - loss: 0.2524 - accuracy: 0.5874 - val_loss: 0.1546 - val_accuracy: 1.0000\n",
            "Epoch 39/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2554 - accuracy: 0.5874 - val_loss: 0.1209 - val_accuracy: 0.9630\n",
            "Epoch 40/100\n",
            "27/27 [==============================] - 8s 296ms/step - loss: 0.2528 - accuracy: 0.5874 - val_loss: 0.0949 - val_accuracy: 1.0000\n",
            "Epoch 41/100\n",
            "27/27 [==============================] - 8s 298ms/step - loss: 0.2558 - accuracy: 0.5874 - val_loss: 0.1444 - val_accuracy: 1.0000\n",
            "2/2 [==============================] - 1s 108ms/step - loss: 0.0709 - accuracy: 0.9706\n",
            "Epoch 1/100\n",
            "28/28 [==============================] - 12s 329ms/step - loss: 0.5474 - accuracy: 0.5068 - val_loss: 0.2094 - val_accuracy: 0.9259\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3777 - accuracy: 0.5811 - val_loss: 0.2091 - val_accuracy: 0.8889\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3619 - accuracy: 0.5902 - val_loss: 0.2041 - val_accuracy: 0.8889\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.3206 - accuracy: 0.6073 - val_loss: 0.1464 - val_accuracy: 1.0000\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2963 - accuracy: 0.6096 - val_loss: 0.1712 - val_accuracy: 0.9630\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.3134 - accuracy: 0.6016 - val_loss: 0.2248 - val_accuracy: 0.9630\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2951 - accuracy: 0.6096 - val_loss: 0.1541 - val_accuracy: 0.9630\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 8s 300ms/step - loss: 0.3075 - accuracy: 0.6050 - val_loss: 0.1223 - val_accuracy: 0.9630\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2898 - accuracy: 0.6084 - val_loss: 0.1212 - val_accuracy: 1.0000\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2793 - accuracy: 0.6119 - val_loss: 0.1703 - val_accuracy: 0.9630\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2710 - accuracy: 0.6107 - val_loss: 0.1653 - val_accuracy: 0.9630\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2723 - accuracy: 0.6096 - val_loss: 0.2584 - val_accuracy: 0.9259\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2819 - accuracy: 0.6107 - val_loss: 0.1701 - val_accuracy: 0.9630\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2794 - accuracy: 0.6107 - val_loss: 0.2867 - val_accuracy: 0.9259\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2721 - accuracy: 0.6107 - val_loss: 0.1729 - val_accuracy: 0.9630\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2669 - accuracy: 0.6096 - val_loss: 0.2258 - val_accuracy: 0.9630\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2758 - accuracy: 0.6084 - val_loss: 0.2100 - val_accuracy: 0.9630\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2670 - accuracy: 0.6107 - val_loss: 0.1526 - val_accuracy: 0.9630\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 8s 299ms/step - loss: 0.2735 - accuracy: 0.6119 - val_loss: 0.0751 - val_accuracy: 0.9630\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2658 - accuracy: 0.6107 - val_loss: 0.1199 - val_accuracy: 0.9630\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2620 - accuracy: 0.6119 - val_loss: 0.2568 - val_accuracy: 0.9259\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2620 - accuracy: 0.6107 - val_loss: 0.1646 - val_accuracy: 0.9630\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2556 - accuracy: 0.6119 - val_loss: 0.1187 - val_accuracy: 0.9630\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2495 - accuracy: 0.6119 - val_loss: 0.1558 - val_accuracy: 0.9630\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2520 - accuracy: 0.6119 - val_loss: 0.1101 - val_accuracy: 0.9630\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2489 - accuracy: 0.6107 - val_loss: 0.1983 - val_accuracy: 0.9630\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2580 - accuracy: 0.6119 - val_loss: 0.3418 - val_accuracy: 0.8519\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2540 - accuracy: 0.6119 - val_loss: 0.2230 - val_accuracy: 0.9259\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2456 - accuracy: 0.6119 - val_loss: 0.1215 - val_accuracy: 0.9630\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 8s 297ms/step - loss: 0.2530 - accuracy: 0.6107 - val_loss: 0.1305 - val_accuracy: 0.9630\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 8s 296ms/step - loss: 0.2443 - accuracy: 0.6119 - val_loss: 0.1585 - val_accuracy: 0.9630\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2477 - accuracy: 0.6119 - val_loss: 0.1570 - val_accuracy: 0.9630\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 8s 295ms/step - loss: 0.2421 - accuracy: 0.6119 - val_loss: 0.0927 - val_accuracy: 0.9630\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 8s 298ms/step - loss: 0.2429 - accuracy: 0.6119 - val_loss: 0.1290 - val_accuracy: 0.9259\n",
            "2/2 [==============================] - 1s 110ms/step - loss: 0.1391 - accuracy: 0.9118\n",
            "[0.9823529362678528, 0.970588219165802, 0.9823529243469238, 0.9764705777168274, 0.9647058606147766]\n",
            "[0.9789437689969607, 0.9654841563514089, 0.9796508230180756, 0.9736615994756604, 0.9603545628048348]\n",
            "Mean test accuracy is 0.98, mean test f1 score is 0.97, max test accuracy is 0.98, max test f1 score is 0.98, min test accuracy is 0.96, min test f1 score is 0.96, std of test accuracy is 0.01, std of test f1 score is 0.01\n",
            "Time elapsed through all process: 8976.75, sec\n"
          ]
        }
      ],
      "source": [
        "t = time.time()\n",
        "# ---------- Parameters ----------------\n",
        "augmentation_enable = True\n",
        "normalize_inputs_enable = True\n",
        "num_folds = 5\n",
        "kfold = StratifiedKFold(n_splits=num_folds, shuffle=True, random_state = None) # random_state = 1 ile split run'dan run'a sabit.\n",
        "test_accuracy_per_run = []\n",
        "f1_score_per_run = []\n",
        "epoch_number = 100\n",
        "batch_size = 32\n",
        "dense_size = 64\n",
        "dropout_prob_dense = 0.5\n",
        "repeat_of_mixup = 5\n",
        "number_of_repeat = 5\n",
        "lstm_dropout_rate = 0.3\n",
        "unit_number_of_lstm = 16\n",
        "dense_unit_of_lstm_function = 256\n",
        "decoder_dense = 256\n",
        "\n",
        "\n",
        "for repeat_run_number in range(number_of_repeat):\n",
        "  test_accuracy_per_fold = []\n",
        "  f1_score_per_fold = []\n",
        "  if repeat_run_number > 0:\n",
        "    del range_doppler_concat_shuffle_test\n",
        "    del spectrogram_concat_shuffle_test\n",
        "    del range_doppler_augmented_image\n",
        "    del range_doppler_concat_shuffle_train\n",
        "    del spectrogram_concat_shuffle_train\n",
        "    del spectrogram_augmented_image\n",
        "   \n",
        "  for randomlist_for_train_indx, randomlist_for_test_indx in kfold.split(range_doppler_concat_shuffle,range_doppler_concat_label_shuffle):   \n",
        "    gc.collect()\n",
        "    K.clear_session()\n",
        "    \n",
        "    # test data\n",
        "    range_doppler_concat_shuffle_test = range_doppler_concat_shuffle[randomlist_for_test_indx,:,:,:]\n",
        "    spectrogram_concat_shuffle_test = spectrogram_concat_shuffle[randomlist_for_test_indx,:,:,:]\n",
        "    range_doppler_concat_label_shuffle_test = range_doppler_concat_label_shuffle[randomlist_for_test_indx,:]\n",
        "    #train data\n",
        "    range_doppler_concat_shuffle_train = range_doppler_concat_shuffle[randomlist_for_train_indx,:,:,:]\n",
        "    spectrogram_concat_shuffle_train = spectrogram_concat_shuffle[randomlist_for_train_indx,:,:,:]\n",
        "    spectrogram_concat_label_shuffle_train = spectrogram_concat_label_shuffle[randomlist_for_train_indx,:]\n",
        "      # ---------------- MixUp Augmentation ----------------\n",
        "    (spectrogram_augmented_image,range_doppler_augmented_image,spectrogram_concat_label_shuffle_concat,\\\n",
        "     validation_spectrogram,validation_range_doppler, spectrogram_validation_labels)  =\\\n",
        "      split_and_augmentation_of_training(spectrogram_concat_shuffle_train,range_doppler_concat_shuffle_train,\\\n",
        "                                         spectrogram_concat_label_shuffle_train,\\\n",
        "                                         repeat_of_mixup, augmentation_enable)\n",
        "    \n",
        "    # ---------------- Neural Network Architecture ----------------\n",
        "\n",
        "\n",
        "\n",
        "    def lstm_encoder_network_1(input_shape):\n",
        "        input = Input(shape=input_shape)\n",
        "        x = Bidirectional(LSTM(unit_number_of_lstm, return_sequences=True, dropout = 0.5))(input)\n",
        "        x = Flatten()(x)\n",
        "        x = BatchNormalization()(x)\n",
        "        x = Dropout(0.5)(x)\n",
        "        x = Dense(dense_unit_of_lstm_function)(x)\n",
        "        x = BatchNormalization()(x)\n",
        "        x = Activation('LeakyReLU')(x)\n",
        "        x = Dropout(dropout_prob_dense)(x)\n",
        "        return Model(input, x)\n",
        "\n",
        "\n",
        "    def decoder_for_concat(input_shape):\n",
        "      input = Input(shape=input_shape)\n",
        "      x = Dense(decoder_dense)(input)\n",
        "      # x = BatchNormalization()(x)\n",
        "      x = Activation('LeakyReLU')(x)\n",
        "      x = Dropout(0.3)(x)\n",
        "      x = Dense(dense_size)(x)\n",
        "      # x = BatchNormalization()(x)\n",
        "      x = Activation('LeakyReLU')(x)\n",
        "      x = Dropout(dropout_prob_dense)(x)\n",
        "      x = Dense(1, activation=\"sigmoid\")(x)\n",
        "      return Model(input, x)\n",
        "\n",
        "    # input_shape = range_doppler_concat_shuffle.shape[1:]\n",
        "    # base_network_range_doppler = encoder_for_range_doppler(input_shape)\n",
        "    # range_doppler_input  = Input(shape=input_shape)\n",
        "    # processed_range_doppler  = base_network_range_doppler(range_doppler_input)\n",
        "\n",
        "    input_shape = range_doppler_concat_shuffle.shape[1:3]\n",
        "    base_network_lstm = lstm_encoder_network_1(input_shape)\n",
        "    range_doppler_input  = Input(shape=input_shape)\n",
        "    processed_lstm  = base_network_lstm(range_doppler_input)\n",
        "\n",
        "    base_decoder_network = decoder_for_concat((processed_lstm.shape[1]))\n",
        "    out = base_decoder_network(processed_lstm)\n",
        "\n",
        "    model = Model(inputs=[range_doppler_input], outputs=[out]) \n",
        "    if repeat_run_number == 0:\n",
        "      print(base_network_lstm.summary())\n",
        "      print(model.summary())\n",
        "    # ---------------- Compile and Fit ----------------\n",
        "    model.compile(loss='binary_crossentropy',\n",
        "                  optimizer='adam',\n",
        "                  metrics=['accuracy'])\n",
        "    \n",
        "    earlyStopping = EarlyStopping(monitor='val_loss', patience=15, verbose=0,restore_best_weights=True, mode='min')\n",
        "    history = model.fit((range_doppler_augmented_image),(spectrogram_concat_label_shuffle_concat),\n",
        "                    epochs=epoch_number,\n",
        "                    batch_size=batch_size,\n",
        "                    shuffle = True,\n",
        "                    callbacks=[earlyStopping],\n",
        "                    validation_data = ((validation_range_doppler) , (spectrogram_validation_labels)))\n",
        "    test_loss, test_accuracy  = model.evaluate([range_doppler_concat_shuffle_test],\\\n",
        "                                               [range_doppler_concat_label_shuffle_test],\n",
        "                  batch_size=batch_size)\n",
        "    gc.collect()\n",
        "    # ---------------- Get Test Results ----------------\n",
        "    y_test_predicted = model.predict((range_doppler_concat_shuffle_test), batch_size=batch_size)\n",
        "    # ----- Binarize y_test_predicted values -----\n",
        "    y_test_predicted_binary = np.zeros(y_test_predicted.size)\n",
        "    for ii in range(y_test_predicted.size):\n",
        "      if y_test_predicted[ii] < 0.5:\n",
        "        y_test_predicted_binary[ii] = 0\n",
        "      else:\n",
        "        y_test_predicted_binary[ii] = 1\n",
        "    \n",
        "    test_precision, test_recall, test_f1_score, support = precision_recall_fscore_support(range_doppler_concat_label_shuffle_test, y_test_predicted_binary, average='macro')\n",
        "\n",
        "    test_accuracy_per_fold.append(test_accuracy)\n",
        "    f1_score_per_fold.append(test_f1_score)\n",
        "    del model\n",
        "  test_accuracy_per_run.append(sum(test_accuracy_per_fold)/num_folds)\n",
        "  f1_score_per_run.append(sum(f1_score_per_fold)/num_folds)\n",
        "  print(test_accuracy_per_run)\n",
        "  print(f1_score_per_run)\n",
        "print(f'Mean test accuracy is {\"{:.2f}\".format(sum(test_accuracy_per_run)/number_of_repeat)}, mean test f1 score is {\"{:.2f}\".format(sum(f1_score_per_run)/number_of_repeat)}, \\\n",
        "max test accuracy is {\"{:.2f}\".format(max(test_accuracy_per_run))}, max test f1 score is {\"{:.2f}\".format(max(f1_score_per_run))}, \\\n",
        "min test accuracy is {\"{:.2f}\".format(min(test_accuracy_per_run))}, min test f1 score is {\"{:.2f}\".format(min(f1_score_per_run))}, \\\n",
        "std of test accuracy is {\"{:.2f}\".format(np.std(test_accuracy_per_run, axis=0))}, std of test f1 score is {\"{:.2f}\".format(np.std(f1_score_per_run, axis=0))}')\n",
        "elapsed = time.time() - t\n",
        "print(f'Time elapsed through all process: {\"{:.2f}\".format(elapsed)}, sec')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "0N7v4PDP4Dv2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        },
        "outputId": "decc2e43-8d66-4cc4-a9f1-57e0105b31a1"
      },
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-1-5a23d0e0df1c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'Mean test accuracy is {\"{:.3f}\".format(sum(test_accuracy_per_run)/number_of_repeat)}, mean test f1 score is {\"{:.3f}\".format(sum(f1_score_per_run)/number_of_repeat)}, max test accuracy is {\"{:.3f}\".format(max(test_accuracy_per_run))}, max test f1 score is {\"{:.3f}\".format(max(f1_score_per_run))}, min test accuracy is {\"{:.3f}\".format(min(test_accuracy_per_run))}, min test f1 score is {\"{:.3f}\".format(min(f1_score_per_run))}, std of test accuracy is {\"{:.3f}\".format(np.std(test_accuracy_per_run, axis=0))}, std of test f1 score is {\"{:.3f}\".format(np.std(f1_score_per_run, axis=0))}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;31m# print(f'Time elapsed through all process: {\"{:.3f}\".format(elapsed)}, sec')\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'test_accuracy_per_run' is not defined"
          ]
        }
      ],
      "source": [
        "print(f'Mean test accuracy is {\"{:.3f}\".format(sum(test_accuracy_per_run)/number_of_repeat)}, mean test f1 score is {\"{:.3f}\".format(sum(f1_score_per_run)/number_of_repeat)}, \\\n",
        "max test accuracy is {\"{:.3f}\".format(max(test_accuracy_per_run))}, max test f1 score is {\"{:.3f}\".format(max(f1_score_per_run))}, \\\n",
        "min test accuracy is {\"{:.3f}\".format(min(test_accuracy_per_run))}, min test f1 score is {\"{:.3f}\".format(min(f1_score_per_run))}, \\\n",
        "std of test accuracy is {\"{:.3f}\".format(np.std(test_accuracy_per_run, axis=0))}, std of test f1 score is {\"{:.3f}\".format(np.std(f1_score_per_run, axis=0))}')\n",
        "# print(f'Time elapsed through all process: {\"{:.3f}\".format(elapsed)}, sec') "
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "machine_shape": "hm",
      "name": "Range_Doppler_LSTM_v2_datalarla_dropout03.ipynb",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}