AP_Pairs.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "uGfyqu0b1ujL"
      },
      "source": [
        "Create Embedding Network"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hdULt7ty1jgQ",
        "outputId": "554b1f3c-ee98-49b4-e9e3-cc059b63c371"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Using:\n",
            "\t• TensorFlow version: 2.12.0\n",
            "\t• tf.keras version: 2.12.0\n",
            "\t• Running on GPU\n"
          ]
        }
      ],
      "source": [
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "import os\n",
        "import time\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "#!pip install tensorflow==2.9.1 # For EfficientNetB0\n",
        "import tensorflow as tf\n",
        "import math\n",
        "\n",
        "import utils\n",
        "\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.preprocessing import image\n",
        "from tensorflow.keras.applications import vgg16, imagenet_utils\n",
        "from tensorflow.keras.optimizers import Adam, RMSprop\n",
        "from tensorflow.keras.regularizers import l2\n",
        "from tensorflow.keras.metrics import binary_crossentropy\n",
        "from tensorflow.keras.layers import Dense, Flatten, Input, Lambda, Dropout, BatchNormalization, GlobalAveragePooling2D\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
        "\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, accuracy_score, precision_score, recall_score\n",
        "import itertools\n",
        "\n",
        "from skimage import exposure\n",
        "\n",
        "import logging\n",
        "from tensorflow.keras.models import load_model\n",
        "\n",
        "logger = tf.get_logger()\n",
        "logger.setLevel(logging.ERROR)\n",
        "\n",
        "print('Using:')\n",
        "print('\\t\\u2022 TensorFlow version:', tf.__version__)\n",
        "print('\\t\\u2022 tf.keras version:', tf.keras.__version__)\n",
        "print('\\t\\u2022 Running on GPU' if tf.test.is_gpu_available() else '\\t\\u2022 GPU device not found. Running on CPU')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "UOYYsI6o157S"
      },
      "source": [
        "Load Dataset (Using 150x150)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zIrlgW2m1tbH",
        "outputId": "baf957da-f80d-42f9-e804-29f4732d6892"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 1372 images belonging to 4 classes.\n",
            "Found 40 images belonging to 4 classes.\n",
            "The dataset contains:\n",
            "• 1372 training images\n",
            "• 40 validation images\n",
            "\n",
            "The training set contains:\n",
            "• 343 glioma images\n",
            "• 343 meningioma images\n",
            "• 343 normal images\n",
            "• 343 pituitary images\n",
            "\n",
            "The validation set contains:\n",
            "• 10 glioma images\n",
            "• 10 meningioma images\n",
            "• 10 normal images\n",
            "• 10 pituitary images\n"
          ]
        }
      ],
      "source": [
        "\"\"\" load the datasets \"\"\"\n",
        "\n",
        "base_dir = os.path.join('dataset', 'pretrain4w')\n",
        "\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "val_dir = os.path.join(base_dir, 'validation')\n",
        "test_dir = os.path.join(base_dir, 'test')\n",
        "\n",
        "train_glioma_dir = os.path.join(train_dir, 'glioma_tumor')\n",
        "train_meningioma_dir = os.path.join(train_dir, 'meningioma_tumor')\n",
        "train_normal_dir = os.path.join(train_dir, 'no_tumor')\n",
        "train_pituitary_dir = os.path.join(train_dir, 'pituitary_tumor')\n",
        "\n",
        "val_glioma_dir = os.path.join(val_dir, 'glioma_tumor')\n",
        "val_meningioma_dir = os.path.join(val_dir, 'meningioma_tumor')\n",
        "val_normal_dir = os.path.join(val_dir, 'no_tumor')\n",
        "val_pituitary_dir = os.path.join(val_dir, 'pituitary_tumor')\n",
        "\n",
        "INPUT_SIZE = 150\n",
        "BATCH_SIZE = 16\n",
        "\n",
        "\"\"\" Investigate train - val datasets \"\"\"\n",
        "\n",
        "train_batches = ImageDataGenerator(rescale = 1 / 255.).flow_from_directory(train_dir,\n",
        "                                                         target_size=(INPUT_SIZE,INPUT_SIZE),\n",
        "                                                         class_mode='categorical',\n",
        "                                                         shuffle=True,\n",
        "                                                         seed=42,\n",
        "                                                         batch_size=BATCH_SIZE)\n",
        "\n",
        "val_batches = ImageDataGenerator(rescale = 1 / 255.).flow_from_directory(val_dir,\n",
        "                                                         target_size=(INPUT_SIZE,INPUT_SIZE),\n",
        "                                                         class_mode='categorical',\n",
        "                                                         shuffle=True,\n",
        "                                                         seed=42,\n",
        "                                                         batch_size=BATCH_SIZE)\n",
        "\n",
        "num_glioma_train = int(len(os.listdir(train_glioma_dir)))\n",
        "num_meningioma_train = int(len(os.listdir(train_meningioma_dir)))\n",
        "num_normal_train = int(len(os.listdir(train_normal_dir)))\n",
        "num_pituitary_train = int(len(os.listdir(train_pituitary_dir)))\n",
        "\n",
        "num_glioma_val = int(len(os.listdir(val_glioma_dir)))\n",
        "num_meningioma_val = int(len(os.listdir(val_meningioma_dir)))\n",
        "num_normal_val = int(len(os.listdir(val_normal_dir)))\n",
        "num_pituitary_val = int(len(os.listdir(val_pituitary_dir)))\n",
        "\n",
        "print('The dataset contains:')\n",
        "print('\\u2022 {} training images'.format(num_glioma_train + num_meningioma_train + num_normal_train + num_pituitary_train))\n",
        "print('\\u2022 {} validation images'.format(num_glioma_val + num_meningioma_val + num_normal_val + num_pituitary_val))\n",
        "\n",
        "print('\\nThe training set contains:')\n",
        "print('\\u2022 {} glioma images'.format(num_glioma_train))\n",
        "print('\\u2022 {} meningioma images'.format(num_meningioma_train))\n",
        "print('\\u2022 {} normal images'.format(num_normal_train))\n",
        "print('\\u2022 {} pituitary images'.format(num_pituitary_train))\n",
        "\n",
        "print('\\nThe validation set contains:')\n",
        "print('\\u2022 {} glioma images'.format(num_glioma_val))\n",
        "print('\\u2022 {} meningioma images'.format(num_meningioma_val))\n",
        "print('\\u2022 {} normal images'.format(num_normal_val))\n",
        "print('\\u2022 {} pituitary images'.format(num_pituitary_val))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "moIDBCAz3mP8",
        "outputId": "5057ad04-a538-465c-9d95-bd12492357d6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "58889256/58889256 [==============================] - 0s 0us/step\n",
            "Number of layers in the base model:  19\n",
            "Model: \"vgg16\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 150, 150, 3)]     0         \n",
            "                                                                 \n",
            " block1_conv1 (Conv2D)       (None, 150, 150, 64)      1792      \n",
            "                                                                 \n",
            " block1_conv2 (Conv2D)       (None, 150, 150, 64)      36928     \n",
            "                                                                 \n",
            " block1_pool (MaxPooling2D)  (None, 75, 75, 64)        0         \n",
            "                                                                 \n",
            " block2_conv1 (Conv2D)       (None, 75, 75, 128)       73856     \n",
            "                                                                 \n",
            " block2_conv2 (Conv2D)       (None, 75, 75, 128)       147584    \n",
            "                                                                 \n",
            " block2_pool (MaxPooling2D)  (None, 37, 37, 128)       0         \n",
            "                                                                 \n",
            " block3_conv1 (Conv2D)       (None, 37, 37, 256)       295168    \n",
            "                                                                 \n",
            " block3_conv2 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_conv3 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_pool (MaxPooling2D)  (None, 18, 18, 256)       0         \n",
            "                                                                 \n",
            " block4_conv1 (Conv2D)       (None, 18, 18, 512)       1180160   \n",
            "                                                                 \n",
            " block4_conv2 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_conv3 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_pool (MaxPooling2D)  (None, 9, 9, 512)         0         \n",
            "                                                                 \n",
            " block5_conv1 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv2 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv3 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_pool (MaxPooling2D)  (None, 4, 4, 512)         0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 14,714,688\n",
            "Trainable params: 14,714,688\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 150, 150, 3)]     0         \n",
            "                                                                 \n",
            " block1_conv1 (Conv2D)       (None, 150, 150, 64)      1792      \n",
            "                                                                 \n",
            " block1_conv2 (Conv2D)       (None, 150, 150, 64)      36928     \n",
            "                                                                 \n",
            " block1_pool (MaxPooling2D)  (None, 75, 75, 64)        0         \n",
            "                                                                 \n",
            " block2_conv1 (Conv2D)       (None, 75, 75, 128)       73856     \n",
            "                                                                 \n",
            " block2_conv2 (Conv2D)       (None, 75, 75, 128)       147584    \n",
            "                                                                 \n",
            " block2_pool (MaxPooling2D)  (None, 37, 37, 128)       0         \n",
            "                                                                 \n",
            " block3_conv1 (Conv2D)       (None, 37, 37, 256)       295168    \n",
            "                                                                 \n",
            " block3_conv2 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_conv3 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_pool (MaxPooling2D)  (None, 18, 18, 256)       0         \n",
            "                                                                 \n",
            " block4_conv1 (Conv2D)       (None, 18, 18, 512)       1180160   \n",
            "                                                                 \n",
            " block4_conv2 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_conv3 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_pool (MaxPooling2D)  (None, 9, 9, 512)         0         \n",
            "                                                                 \n",
            " block5_conv1 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv2 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv3 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_pool (MaxPooling2D)  (None, 4, 4, 512)         0         \n",
            "                                                                 \n",
            " global_average_pooling2d (G  (None, 512)              0         \n",
            " lobalAveragePooling2D)                                          \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 512)               0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 512)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 4)                 2052      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 14,716,740\n",
            "Trainable params: 14,716,740\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "MODEL_FNAME = 'embedding_network.h5'\n",
        "\n",
        "tf.compat.v1.reset_default_graph()\n",
        "\n",
        "tf.random.set_seed(0)\n",
        "tf.keras.utils.set_random_seed(0)\n",
        "\n",
        "\n",
        "\"\"\" Create the model \"\"\"\n",
        "    \n",
        "base_model = vgg16.VGG16(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.1161 - val_accuracy: 0.9750 - Test: 0.94557\n",
        "#base_model = EfficientNetB0(weights='imagenet',include_top=False,input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0312 - accuracy: 0.9920 - val_loss: 0.1799 - val_accuracy: 0.9250 - Test: 0.933673\n",
        "#base_model = Xception(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0335 - accuracy: 0.9978 - val_loss: 0.3005 - val_accuracy: 0.9000\n",
        "#base_model = InceptionV3(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0912 - accuracy: 0.9730 - val_loss: 0.1897 - val_accuracy: 0.9250\n",
        "#base_model = InceptionResNetV2(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0063 - accuracy: 0.9978 - val_loss: 0.2221 - val_accuracy: 0.9250\n",
        "#base_model = ResNet50(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0028 - accuracy: 0.9985 - val_loss: 0.1825 - val_accuracy: 0.9250\n",
        "#base_model = vgg19.VGG19(weights='imagenet', include_top=False, input_shape=(INPUT_SIZE,INPUT_SIZE,3)) # loss: 0.0149 - accuracy: 0.9956 - val_loss: 0.1836 - val_accuracy: 0.9750 - Test: 0.930272\n",
        "\n",
        "print(\"Number of layers in the base model: \", len(base_model.layers))\n",
        "    \n",
        "base_model.trainable = True\n",
        "    \n",
        "base_model.summary()\n",
        "    \n",
        "last_output = base_model.output\n",
        "\n",
        "x = GlobalAveragePooling2D()(last_output)\n",
        "x = Dropout(rate=0.5)(x)\n",
        "\n",
        "x = Flatten()(x)\n",
        "x = Dense(4, activation='softmax')(x)\n",
        "    \n",
        "embedding_network = Model(inputs=[base_model.input], outputs=[x])\n",
        "    \n",
        "optimizer = Adam(learning_rate=0.00001) \n",
        "embedding_network.compile(loss='categorical_crossentropy', optimizer=optimizer, metrics=[\"accuracy\"])\n",
        "embedding_network.summary()\n",
        "    \n",
        "\"\"\" callbacks \"\"\"\n",
        "    \n",
        "#early_stopping = EarlyStopping(monitor='val_loss', patience=10)\n",
        "    \n",
        "checkpointer = ModelCheckpoint(filepath='embedding_network.h5', verbose=1, \n",
        "                                   save_best_only=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "JkleuxdI3yiH",
        "outputId": "cd15333e-e1d9-4f16-ff0e-589668717eb6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.1561 - accuracy: 0.4796\n",
            "Epoch 1: val_loss improved from inf to 0.76707, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 37s 180ms/step - loss: 1.1561 - accuracy: 0.4796 - val_loss: 0.7671 - val_accuracy: 0.7250\n",
            "Epoch 2/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.7317 - accuracy: 0.7230\n",
            "Epoch 2: val_loss improved from 0.76707 to 0.53831, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 10s 116ms/step - loss: 0.7317 - accuracy: 0.7230 - val_loss: 0.5383 - val_accuracy: 0.8000\n",
            "Epoch 3/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.5368 - accuracy: 0.8025\n",
            "Epoch 3: val_loss improved from 0.53831 to 0.50126, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.5368 - accuracy: 0.8025 - val_loss: 0.5013 - val_accuracy: 0.8000\n",
            "Epoch 4/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.4239 - accuracy: 0.8571\n",
            "Epoch 4: val_loss improved from 0.50126 to 0.37906, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.4239 - accuracy: 0.8571 - val_loss: 0.3791 - val_accuracy: 0.9000\n",
            "Epoch 5/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.3331 - accuracy: 0.8812\n",
            "Epoch 5: val_loss improved from 0.37906 to 0.34601, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 13s 147ms/step - loss: 0.3331 - accuracy: 0.8812 - val_loss: 0.3460 - val_accuracy: 0.8750\n",
            "Epoch 6/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.2484 - accuracy: 0.9118\n",
            "Epoch 6: val_loss improved from 0.34601 to 0.32686, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.2484 - accuracy: 0.9118 - val_loss: 0.3269 - val_accuracy: 0.9000\n",
            "Epoch 7/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.2101 - accuracy: 0.9264\n",
            "Epoch 7: val_loss improved from 0.32686 to 0.29101, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.2101 - accuracy: 0.9264 - val_loss: 0.2910 - val_accuracy: 0.9000\n",
            "Epoch 8/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.1806 - accuracy: 0.9373\n",
            "Epoch 8: val_loss did not improve from 0.29101\n",
            "86/86 [==============================] - 11s 122ms/step - loss: 0.1806 - accuracy: 0.9373 - val_loss: 0.3283 - val_accuracy: 0.8750\n",
            "Epoch 9/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.1574 - accuracy: 0.9475\n",
            "Epoch 9: val_loss did not improve from 0.29101\n",
            "86/86 [==============================] - 11s 121ms/step - loss: 0.1574 - accuracy: 0.9475 - val_loss: 0.3520 - val_accuracy: 0.9000\n",
            "Epoch 10/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.1201 - accuracy: 0.9606\n",
            "Epoch 10: val_loss improved from 0.29101 to 0.21346, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.1201 - accuracy: 0.9606 - val_loss: 0.2135 - val_accuracy: 0.9000\n",
            "Epoch 11/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0808 - accuracy: 0.9774\n",
            "Epoch 11: val_loss did not improve from 0.21346\n",
            "86/86 [==============================] - 11s 122ms/step - loss: 0.0808 - accuracy: 0.9774 - val_loss: 0.4683 - val_accuracy: 0.8750\n",
            "Epoch 12/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0719 - accuracy: 0.9752\n",
            "Epoch 12: val_loss did not improve from 0.21346\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.0719 - accuracy: 0.9752 - val_loss: 0.4406 - val_accuracy: 0.9000\n",
            "Epoch 13/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0628 - accuracy: 0.9825\n",
            "Epoch 13: val_loss did not improve from 0.21346\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.0628 - accuracy: 0.9825 - val_loss: 0.5152 - val_accuracy: 0.9250\n",
            "Epoch 14/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0612 - accuracy: 0.9818\n",
            "Epoch 14: val_loss did not improve from 0.21346\n",
            "86/86 [==============================] - 11s 122ms/step - loss: 0.0612 - accuracy: 0.9818 - val_loss: 0.2483 - val_accuracy: 0.9250\n",
            "Epoch 15/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0403 - accuracy: 0.9862\n",
            "Epoch 15: val_loss did not improve from 0.21346\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0403 - accuracy: 0.9862 - val_loss: 0.3837 - val_accuracy: 0.9250\n",
            "Epoch 16/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0453 - accuracy: 0.9847\n",
            "Epoch 16: val_loss improved from 0.21346 to 0.20409, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0453 - accuracy: 0.9847 - val_loss: 0.2041 - val_accuracy: 0.9250\n",
            "Epoch 17/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0357 - accuracy: 0.9913\n",
            "Epoch 17: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0357 - accuracy: 0.9913 - val_loss: 0.2895 - val_accuracy: 0.9250\n",
            "Epoch 18/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0228 - accuracy: 0.9927\n",
            "Epoch 18: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0228 - accuracy: 0.9927 - val_loss: 0.3059 - val_accuracy: 0.9250\n",
            "Epoch 19/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0382 - accuracy: 0.9883\n",
            "Epoch 19: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 10s 121ms/step - loss: 0.0382 - accuracy: 0.9883 - val_loss: 0.3453 - val_accuracy: 0.9000\n",
            "Epoch 20/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0246 - accuracy: 0.9920\n",
            "Epoch 20: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0246 - accuracy: 0.9920 - val_loss: 0.3706 - val_accuracy: 0.9000\n",
            "Epoch 21/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0245 - accuracy: 0.9934\n",
            "Epoch 21: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0245 - accuracy: 0.9934 - val_loss: 0.3394 - val_accuracy: 0.9000\n",
            "Epoch 22/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0320 - accuracy: 0.9883\n",
            "Epoch 22: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 10s 119ms/step - loss: 0.0320 - accuracy: 0.9883 - val_loss: 0.3914 - val_accuracy: 0.9250\n",
            "Epoch 23/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0197 - accuracy: 0.9942\n",
            "Epoch 23: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0197 - accuracy: 0.9942 - val_loss: 0.2796 - val_accuracy: 0.9250\n",
            "Epoch 24/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0195 - accuracy: 0.9956\n",
            "Epoch 24: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0195 - accuracy: 0.9956 - val_loss: 0.5938 - val_accuracy: 0.9250\n",
            "Epoch 25/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0058 - accuracy: 0.9993\n",
            "Epoch 25: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 10s 121ms/step - loss: 0.0058 - accuracy: 0.9993 - val_loss: 0.4822 - val_accuracy: 0.9250\n",
            "Epoch 26/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0673 - accuracy: 0.9752\n",
            "Epoch 26: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 0.0673 - accuracy: 0.9752 - val_loss: 0.2922 - val_accuracy: 0.9250\n",
            "Epoch 27/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0228 - accuracy: 0.9934\n",
            "Epoch 27: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0228 - accuracy: 0.9934 - val_loss: 0.3922 - val_accuracy: 0.9000\n",
            "Epoch 28/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0074 - accuracy: 0.9993\n",
            "Epoch 28: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0074 - accuracy: 0.9993 - val_loss: 0.2867 - val_accuracy: 0.9000\n",
            "Epoch 29/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0080 - accuracy: 0.9978\n",
            "Epoch 29: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.0080 - accuracy: 0.9978 - val_loss: 0.2690 - val_accuracy: 0.9250\n",
            "Epoch 30/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0071 - accuracy: 0.9971\n",
            "Epoch 30: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0071 - accuracy: 0.9971 - val_loss: 0.2111 - val_accuracy: 0.9250\n",
            "Epoch 31/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0075 - accuracy: 0.9978\n",
            "Epoch 31: val_loss did not improve from 0.20409\n",
            "86/86 [==============================] - 10s 119ms/step - loss: 0.0075 - accuracy: 0.9978 - val_loss: 0.4415 - val_accuracy: 0.9250\n",
            "Epoch 32/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0214 - accuracy: 0.9934\n",
            "Epoch 32: val_loss improved from 0.20409 to 0.15018, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0214 - accuracy: 0.9934 - val_loss: 0.1502 - val_accuracy: 0.9500\n",
            "Epoch 33/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0084 - accuracy: 0.9993\n",
            "Epoch 33: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0084 - accuracy: 0.9993 - val_loss: 0.5673 - val_accuracy: 0.9000\n",
            "Epoch 34/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0043 - accuracy: 0.9985\n",
            "Epoch 34: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0043 - accuracy: 0.9985 - val_loss: 0.4531 - val_accuracy: 0.8500\n",
            "Epoch 35/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0267 - accuracy: 0.9934\n",
            "Epoch 35: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.0267 - accuracy: 0.9934 - val_loss: 0.2887 - val_accuracy: 0.9250\n",
            "Epoch 36/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0016 - accuracy: 1.0000\n",
            "Epoch 36: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0016 - accuracy: 1.0000 - val_loss: 0.3180 - val_accuracy: 0.9250\n",
            "Epoch 37/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0030 - accuracy: 0.9993\n",
            "Epoch 37: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0030 - accuracy: 0.9993 - val_loss: 0.2988 - val_accuracy: 0.9000\n",
            "Epoch 38/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0234 - accuracy: 0.9920\n",
            "Epoch 38: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 0.0234 - accuracy: 0.9920 - val_loss: 0.2920 - val_accuracy: 0.9000\n",
            "Epoch 39/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0024 - accuracy: 1.0000\n",
            "Epoch 39: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0024 - accuracy: 1.0000 - val_loss: 0.3905 - val_accuracy: 0.9000\n",
            "Epoch 40/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0015 - accuracy: 1.0000\n",
            "Epoch 40: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.4280 - val_accuracy: 0.9000\n",
            "Epoch 41/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.4194e-04 - accuracy: 1.0000\n",
            "Epoch 41: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 10s 119ms/step - loss: 9.4194e-04 - accuracy: 1.0000 - val_loss: 0.3882 - val_accuracy: 0.9000\n",
            "Epoch 42/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.1910e-04 - accuracy: 1.0000\n",
            "Epoch 42: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 9.1910e-04 - accuracy: 1.0000 - val_loss: 0.3661 - val_accuracy: 0.9000\n",
            "Epoch 43/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.0847e-04 - accuracy: 1.0000\n",
            "Epoch 43: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 5.0847e-04 - accuracy: 1.0000 - val_loss: 0.4099 - val_accuracy: 0.9000\n",
            "Epoch 44/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0011 - accuracy: 1.0000\n",
            "Epoch 44: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.6148 - val_accuracy: 0.9250\n",
            "Epoch 45/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0048 - accuracy: 0.9978\n",
            "Epoch 45: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 10s 120ms/step - loss: 0.0048 - accuracy: 0.9978 - val_loss: 0.6651 - val_accuracy: 0.9000\n",
            "Epoch 46/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0377 - accuracy: 0.9898\n",
            "Epoch 46: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0377 - accuracy: 0.9898 - val_loss: 0.3369 - val_accuracy: 0.9500\n",
            "Epoch 47/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0099 - accuracy: 0.9964\n",
            "Epoch 47: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0099 - accuracy: 0.9964 - val_loss: 0.2021 - val_accuracy: 0.9500\n",
            "Epoch 48/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0188 - accuracy: 0.9934\n",
            "Epoch 48: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 0.0188 - accuracy: 0.9934 - val_loss: 0.3959 - val_accuracy: 0.9250\n",
            "Epoch 49/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0152 - accuracy: 0.9964\n",
            "Epoch 49: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0152 - accuracy: 0.9964 - val_loss: 0.3904 - val_accuracy: 0.9000\n",
            "Epoch 50/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0101 - accuracy: 0.9978\n",
            "Epoch 50: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0101 - accuracy: 0.9978 - val_loss: 0.2539 - val_accuracy: 0.9500\n",
            "Epoch 51/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0270 - accuracy: 0.9920\n",
            "Epoch 51: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0270 - accuracy: 0.9920 - val_loss: 0.1944 - val_accuracy: 0.9250\n",
            "Epoch 52/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0147 - accuracy: 0.9964\n",
            "Epoch 52: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 122ms/step - loss: 0.0147 - accuracy: 0.9964 - val_loss: 0.3286 - val_accuracy: 0.9000\n",
            "Epoch 53/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0011 - accuracy: 1.0000\n",
            "Epoch 53: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.2314 - val_accuracy: 0.9250\n",
            "Epoch 54/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 8.1244e-04 - accuracy: 1.0000\n",
            "Epoch 54: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 8.1244e-04 - accuracy: 1.0000 - val_loss: 0.2506 - val_accuracy: 0.9250\n",
            "Epoch 55/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0015 - accuracy: 1.0000\n",
            "Epoch 55: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.3484 - val_accuracy: 0.9000\n",
            "Epoch 56/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 7.6822e-04 - accuracy: 1.0000\n",
            "Epoch 56: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 7.6822e-04 - accuracy: 1.0000 - val_loss: 0.3435 - val_accuracy: 0.9000\n",
            "Epoch 57/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.3606e-04 - accuracy: 1.0000\n",
            "Epoch 57: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 5.3606e-04 - accuracy: 1.0000 - val_loss: 0.3585 - val_accuracy: 0.9250\n",
            "Epoch 58/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 2.8080e-04 - accuracy: 1.0000\n",
            "Epoch 58: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 2.8080e-04 - accuracy: 1.0000 - val_loss: 0.3141 - val_accuracy: 0.9000\n",
            "Epoch 59/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.7745e-04 - accuracy: 1.0000\n",
            "Epoch 59: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 1.7745e-04 - accuracy: 1.0000 - val_loss: 0.3258 - val_accuracy: 0.9250\n",
            "Epoch 60/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0020 - accuracy: 1.0000\n",
            "Epoch 60: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0020 - accuracy: 1.0000 - val_loss: 0.4791 - val_accuracy: 0.9250\n",
            "Epoch 61/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0056 - accuracy: 0.9971\n",
            "Epoch 61: val_loss did not improve from 0.15018\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0056 - accuracy: 0.9971 - val_loss: 0.4714 - val_accuracy: 0.9250\n",
            "Epoch 62/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0081 - accuracy: 0.9985\n",
            "Epoch 62: val_loss improved from 0.15018 to 0.10925, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 12s 140ms/step - loss: 0.0081 - accuracy: 0.9985 - val_loss: 0.1093 - val_accuracy: 0.9500\n",
            "Epoch 63/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0016 - accuracy: 1.0000\n",
            "Epoch 63: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 12s 143ms/step - loss: 0.0016 - accuracy: 1.0000 - val_loss: 0.2221 - val_accuracy: 0.9250\n",
            "Epoch 64/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0017 - accuracy: 1.0000\n",
            "Epoch 64: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 12s 135ms/step - loss: 0.0017 - accuracy: 1.0000 - val_loss: 0.3176 - val_accuracy: 0.9250\n",
            "Epoch 65/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 2.9295e-04 - accuracy: 1.0000\n",
            "Epoch 65: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 2.9295e-04 - accuracy: 1.0000 - val_loss: 0.2436 - val_accuracy: 0.9250\n",
            "Epoch 66/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0065 - accuracy: 0.9978\n",
            "Epoch 66: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0065 - accuracy: 0.9978 - val_loss: 1.3482 - val_accuracy: 0.8750\n",
            "Epoch 67/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0541 - accuracy: 0.9840\n",
            "Epoch 67: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0541 - accuracy: 0.9840 - val_loss: 0.1684 - val_accuracy: 0.9250\n",
            "Epoch 68/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0231 - accuracy: 0.9934\n",
            "Epoch 68: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 130ms/step - loss: 0.0231 - accuracy: 0.9934 - val_loss: 0.3434 - val_accuracy: 0.9250\n",
            "Epoch 69/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0028 - accuracy: 0.9993\n",
            "Epoch 69: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0028 - accuracy: 0.9993 - val_loss: 0.4287 - val_accuracy: 0.9250\n",
            "Epoch 70/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0038 - accuracy: 0.9993\n",
            "Epoch 70: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0038 - accuracy: 0.9993 - val_loss: 0.5005 - val_accuracy: 0.9250\n",
            "Epoch 71/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0039 - accuracy: 0.9985\n",
            "Epoch 71: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0039 - accuracy: 0.9985 - val_loss: 0.2234 - val_accuracy: 0.9500\n",
            "Epoch 72/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 4.3308e-04 - accuracy: 1.0000\n",
            "Epoch 72: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 4.3308e-04 - accuracy: 1.0000 - val_loss: 0.4271 - val_accuracy: 0.9250\n",
            "Epoch 73/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.2328e-04 - accuracy: 1.0000\n",
            "Epoch 73: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 131ms/step - loss: 3.2328e-04 - accuracy: 1.0000 - val_loss: 0.3658 - val_accuracy: 0.9250\n",
            "Epoch 74/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 2.4773e-04 - accuracy: 1.0000\n",
            "Epoch 74: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 2.4773e-04 - accuracy: 1.0000 - val_loss: 0.2978 - val_accuracy: 0.9500\n",
            "Epoch 75/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 2.2880e-04 - accuracy: 1.0000\n",
            "Epoch 75: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 2.2880e-04 - accuracy: 1.0000 - val_loss: 0.3363 - val_accuracy: 0.9250\n",
            "Epoch 76/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.4023e-04 - accuracy: 1.0000\n",
            "Epoch 76: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 1.4023e-04 - accuracy: 1.0000 - val_loss: 0.3588 - val_accuracy: 0.9250\n",
            "Epoch 77/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 8.2563e-05 - accuracy: 1.0000\n",
            "Epoch 77: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 8.2563e-05 - accuracy: 1.0000 - val_loss: 0.3547 - val_accuracy: 0.9250\n",
            "Epoch 78/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.3979e-05 - accuracy: 1.0000\n",
            "Epoch 78: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 9.3979e-05 - accuracy: 1.0000 - val_loss: 0.3685 - val_accuracy: 0.9250\n",
            "Epoch 79/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 6.3850e-04 - accuracy: 1.0000\n",
            "Epoch 79: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 6.3850e-04 - accuracy: 1.0000 - val_loss: 0.4795 - val_accuracy: 0.9250\n",
            "Epoch 80/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0112 - accuracy: 0.9978\n",
            "Epoch 80: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0112 - accuracy: 0.9978 - val_loss: 0.3646 - val_accuracy: 0.9250\n",
            "Epoch 81/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0359 - accuracy: 0.9876\n",
            "Epoch 81: val_loss did not improve from 0.10925\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0359 - accuracy: 0.9876 - val_loss: 0.1385 - val_accuracy: 0.9250\n",
            "Epoch 82/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0080 - accuracy: 0.9971\n",
            "Epoch 82: val_loss improved from 0.10925 to 0.10160, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 12s 134ms/step - loss: 0.0080 - accuracy: 0.9971 - val_loss: 0.1016 - val_accuracy: 0.9500\n",
            "Epoch 83/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0169 - accuracy: 0.9949\n",
            "Epoch 83: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0169 - accuracy: 0.9949 - val_loss: 0.2069 - val_accuracy: 0.9500\n",
            "Epoch 84/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0119 - accuracy: 0.9949\n",
            "Epoch 84: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0119 - accuracy: 0.9949 - val_loss: 0.1921 - val_accuracy: 0.9250\n",
            "Epoch 85/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0021 - accuracy: 1.0000\n",
            "Epoch 85: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0021 - accuracy: 1.0000 - val_loss: 0.1212 - val_accuracy: 0.9500\n",
            "Epoch 86/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0059 - accuracy: 0.9985\n",
            "Epoch 86: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0059 - accuracy: 0.9985 - val_loss: 0.2813 - val_accuracy: 0.9500\n",
            "Epoch 87/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0087 - accuracy: 0.9985\n",
            "Epoch 87: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 0.0087 - accuracy: 0.9985 - val_loss: 0.2852 - val_accuracy: 0.9250\n",
            "Epoch 88/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0097 - accuracy: 0.9978\n",
            "Epoch 88: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 0.0097 - accuracy: 0.9978 - val_loss: 0.1203 - val_accuracy: 0.9750\n",
            "Epoch 89/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0022 - accuracy: 0.9993\n",
            "Epoch 89: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0022 - accuracy: 0.9993 - val_loss: 0.1033 - val_accuracy: 0.9500\n",
            "Epoch 90/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 4.3471e-04 - accuracy: 1.0000\n",
            "Epoch 90: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 122ms/step - loss: 4.3471e-04 - accuracy: 1.0000 - val_loss: 0.1160 - val_accuracy: 0.9750\n",
            "Epoch 91/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.5769e-04 - accuracy: 1.0000\n",
            "Epoch 91: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 3.5769e-04 - accuracy: 1.0000 - val_loss: 0.1042 - val_accuracy: 0.9500\n",
            "Epoch 92/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.7830e-04 - accuracy: 1.0000\n",
            "Epoch 92: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 1.7830e-04 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9500\n",
            "Epoch 93/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.3803e-04 - accuracy: 1.0000\n",
            "Epoch 93: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 1.3803e-04 - accuracy: 1.0000 - val_loss: 0.1112 - val_accuracy: 0.9750\n",
            "Epoch 94/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.1361e-04 - accuracy: 1.0000\n",
            "Epoch 94: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 1.1361e-04 - accuracy: 1.0000 - val_loss: 0.1040 - val_accuracy: 0.9750\n",
            "Epoch 95/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 7.9973e-05 - accuracy: 1.0000\n",
            "Epoch 95: val_loss did not improve from 0.10160\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 7.9973e-05 - accuracy: 1.0000 - val_loss: 0.1042 - val_accuracy: 0.9750\n",
            "Epoch 96/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.5170e-05 - accuracy: 1.0000\n",
            "Epoch 96: val_loss improved from 0.10160 to 0.09979, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 130ms/step - loss: 5.5170e-05 - accuracy: 1.0000 - val_loss: 0.0998 - val_accuracy: 0.9750\n",
            "Epoch 97/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 6.1805e-05 - accuracy: 1.0000\n",
            "Epoch 97: val_loss improved from 0.09979 to 0.08677, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 6.1805e-05 - accuracy: 1.0000 - val_loss: 0.0868 - val_accuracy: 0.9500\n",
            "Epoch 98/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.5651e-05 - accuracy: 1.0000\n",
            "Epoch 98: val_loss did not improve from 0.08677\n",
            "86/86 [==============================] - 12s 135ms/step - loss: 3.5651e-05 - accuracy: 1.0000 - val_loss: 0.0869 - val_accuracy: 0.9750\n",
            "Epoch 99/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.9485e-05 - accuracy: 1.0000\n",
            "Epoch 99: val_loss improved from 0.08677 to 0.08633, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 132ms/step - loss: 5.9485e-05 - accuracy: 1.0000 - val_loss: 0.0863 - val_accuracy: 0.9750\n",
            "Epoch 100/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.8404e-05 - accuracy: 1.0000\n",
            "Epoch 100: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 5.8404e-05 - accuracy: 1.0000 - val_loss: 0.1093 - val_accuracy: 0.9500\n",
            "Epoch 101/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 6.1619e-05 - accuracy: 1.0000\n",
            "Epoch 101: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 6.1619e-05 - accuracy: 1.0000 - val_loss: 0.1573 - val_accuracy: 0.9500\n",
            "Epoch 102/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 7.2137e-05 - accuracy: 1.0000\n",
            "Epoch 102: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 7.2137e-05 - accuracy: 1.0000 - val_loss: 0.1633 - val_accuracy: 0.9500\n",
            "Epoch 103/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0023 - accuracy: 0.9993\n",
            "Epoch 103: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0023 - accuracy: 0.9993 - val_loss: 0.3378 - val_accuracy: 0.9000\n",
            "Epoch 104/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0177 - accuracy: 0.9949\n",
            "Epoch 104: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0177 - accuracy: 0.9949 - val_loss: 0.3749 - val_accuracy: 0.9250\n",
            "Epoch 105/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0011 - accuracy: 1.0000\n",
            "Epoch 105: val_loss did not improve from 0.08633\n",
            "86/86 [==============================] - 11s 121ms/step - loss: 0.0011 - accuracy: 1.0000 - val_loss: 0.1152 - val_accuracy: 0.9750\n",
            "Epoch 106/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.1414e-04 - accuracy: 1.0000\n",
            "Epoch 106: val_loss improved from 0.08633 to 0.02977, saving model to embedding_network.h5\n",
            "86/86 [==============================] - 11s 130ms/step - loss: 3.1414e-04 - accuracy: 1.0000 - val_loss: 0.0298 - val_accuracy: 0.9750\n",
            "Epoch 107/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0072 - accuracy: 0.9985\n",
            "Epoch 107: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0072 - accuracy: 0.9985 - val_loss: 0.0437 - val_accuracy: 0.9750\n",
            "Epoch 108/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0185 - accuracy: 0.9956\n",
            "Epoch 108: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0185 - accuracy: 0.9956 - val_loss: 0.5079 - val_accuracy: 0.9250\n",
            "Epoch 109/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0156 - accuracy: 0.9934\n",
            "Epoch 109: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 0.0156 - accuracy: 0.9934 - val_loss: 0.0946 - val_accuracy: 0.9250\n",
            "Epoch 110/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0013 - accuracy: 1.0000\n",
            "Epoch 110: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.1208 - val_accuracy: 0.9500\n",
            "Epoch 111/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0016 - accuracy: 1.0000\n",
            "Epoch 111: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 0.0016 - accuracy: 1.0000 - val_loss: 0.0938 - val_accuracy: 0.9500\n",
            "Epoch 112/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.3184e-04 - accuracy: 1.0000\n",
            "Epoch 112: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 5.3184e-04 - accuracy: 1.0000 - val_loss: 0.0507 - val_accuracy: 0.9750\n",
            "Epoch 113/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.1475e-04 - accuracy: 1.0000\n",
            "Epoch 113: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 1.1475e-04 - accuracy: 1.0000 - val_loss: 0.0625 - val_accuracy: 0.9750\n",
            "Epoch 114/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.1500e-05 - accuracy: 1.0000\n",
            "Epoch 114: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 9.1500e-05 - accuracy: 1.0000 - val_loss: 0.0915 - val_accuracy: 0.9500\n",
            "Epoch 115/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.3003e-05 - accuracy: 1.0000\n",
            "Epoch 115: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 5.3003e-05 - accuracy: 1.0000 - val_loss: 0.1049 - val_accuracy: 0.9500\n",
            "Epoch 116/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 5.4930e-05 - accuracy: 1.0000\n",
            "Epoch 116: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 5.4930e-05 - accuracy: 1.0000 - val_loss: 0.1262 - val_accuracy: 0.9500\n",
            "Epoch 117/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.1146e-04 - accuracy: 1.0000\n",
            "Epoch 117: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 1.1146e-04 - accuracy: 1.0000 - val_loss: 0.1204 - val_accuracy: 0.9500\n",
            "Epoch 118/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 7.8709e-05 - accuracy: 1.0000\n",
            "Epoch 118: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 7.8709e-05 - accuracy: 1.0000 - val_loss: 0.1128 - val_accuracy: 0.9500\n",
            "Epoch 119/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 4.1677e-05 - accuracy: 1.0000\n",
            "Epoch 119: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 4.1677e-05 - accuracy: 1.0000 - val_loss: 0.1186 - val_accuracy: 0.9500\n",
            "Epoch 120/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.2115e-05 - accuracy: 1.0000\n",
            "Epoch 120: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 132ms/step - loss: 3.2115e-05 - accuracy: 1.0000 - val_loss: 0.1288 - val_accuracy: 0.9500\n",
            "Epoch 121/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.3459e-05 - accuracy: 1.0000\n",
            "Epoch 121: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 3.3459e-05 - accuracy: 1.0000 - val_loss: 0.0991 - val_accuracy: 0.9750\n",
            "Epoch 122/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.8280e-05 - accuracy: 1.0000\n",
            "Epoch 122: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 1.8280e-05 - accuracy: 1.0000 - val_loss: 0.0966 - val_accuracy: 0.9750\n",
            "Epoch 123/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 6.8971e-05 - accuracy: 1.0000\n",
            "Epoch 123: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 128ms/step - loss: 6.8971e-05 - accuracy: 1.0000 - val_loss: 0.1732 - val_accuracy: 0.9750\n",
            "Epoch 124/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 3.8610e-05 - accuracy: 1.0000\n",
            "Epoch 124: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 3.8610e-05 - accuracy: 1.0000 - val_loss: 0.1097 - val_accuracy: 0.9500\n",
            "Epoch 125/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0612 - accuracy: 0.9847\n",
            "Epoch 125: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 10s 121ms/step - loss: 0.0612 - accuracy: 0.9847 - val_loss: 0.3003 - val_accuracy: 0.9000\n",
            "Epoch 126/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0236 - accuracy: 0.9927\n",
            "Epoch 126: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 0.0236 - accuracy: 0.9927 - val_loss: 0.2162 - val_accuracy: 0.9500\n",
            "Epoch 127/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0041 - accuracy: 0.9993\n",
            "Epoch 127: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 0.0041 - accuracy: 0.9993 - val_loss: 0.1038 - val_accuracy: 0.9500\n",
            "Epoch 128/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 6.2832e-04 - accuracy: 1.0000\n",
            "Epoch 128: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 6.2832e-04 - accuracy: 1.0000 - val_loss: 0.1199 - val_accuracy: 0.9500\n",
            "Epoch 129/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 4.6157e-04 - accuracy: 1.0000\n",
            "Epoch 129: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 4.6157e-04 - accuracy: 1.0000 - val_loss: 0.0378 - val_accuracy: 0.9750\n",
            "Epoch 130/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.3423e-04 - accuracy: 1.0000\n",
            "Epoch 130: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 1.3423e-04 - accuracy: 1.0000 - val_loss: 0.0702 - val_accuracy: 0.9500\n",
            "Epoch 131/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.9096e-05 - accuracy: 1.0000\n",
            "Epoch 131: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 9.9096e-05 - accuracy: 1.0000 - val_loss: 0.0789 - val_accuracy: 0.9500\n",
            "Epoch 132/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.9798e-04 - accuracy: 1.0000\n",
            "Epoch 132: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 126ms/step - loss: 1.9798e-04 - accuracy: 1.0000 - val_loss: 0.0499 - val_accuracy: 0.9750\n",
            "Epoch 133/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 9.4745e-05 - accuracy: 1.0000\n",
            "Epoch 133: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 9.4745e-05 - accuracy: 1.0000 - val_loss: 0.0644 - val_accuracy: 0.9750\n",
            "Epoch 134/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 1.8789e-04 - accuracy: 1.0000\n",
            "Epoch 134: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 125ms/step - loss: 1.8789e-04 - accuracy: 1.0000 - val_loss: 0.0450 - val_accuracy: 0.9750\n",
            "Epoch 135/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0021 - accuracy: 0.9993\n",
            "Epoch 135: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0021 - accuracy: 0.9993 - val_loss: 0.0483 - val_accuracy: 0.9750\n",
            "Epoch 136/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0252 - accuracy: 0.9927\n",
            "Epoch 136: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 123ms/step - loss: 0.0252 - accuracy: 0.9927 - val_loss: 0.2387 - val_accuracy: 0.9250\n",
            "Epoch 137/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0159 - accuracy: 0.9949\n",
            "Epoch 137: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 129ms/step - loss: 0.0159 - accuracy: 0.9949 - val_loss: 0.1844 - val_accuracy: 0.9250\n",
            "Epoch 138/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0018 - accuracy: 1.0000\n",
            "Epoch 138: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 0.0018 - accuracy: 1.0000 - val_loss: 0.1984 - val_accuracy: 0.9250\n",
            "Epoch 139/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 0.0021 - accuracy: 0.9993\n",
            "Epoch 139: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 124ms/step - loss: 0.0021 - accuracy: 0.9993 - val_loss: 0.0451 - val_accuracy: 0.9500\n",
            "Epoch 140/140\n",
            "86/86 [==============================] - ETA: 0s - loss: 2.6831e-04 - accuracy: 1.0000\n",
            "Epoch 140: val_loss did not improve from 0.02977\n",
            "86/86 [==============================] - 11s 127ms/step - loss: 2.6831e-04 - accuracy: 1.0000 - val_loss: 0.0920 - val_accuracy: 0.9500\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "End of Training, the model is saved to embedding_network.h5\n"
          ]
        }
      ],
      "source": [
        "\"\"\" Train the model \"\"\"\n",
        "    \n",
        "history = embedding_network.fit(\n",
        "    train_batches,\n",
        "    validation_data = val_batches,\n",
        "    epochs = 140,\n",
        "    verbose = 1,\n",
        "    shuffle = True,\n",
        "    callbacks = [checkpointer]#, early_stopping]\n",
        ")\n",
        "    \n",
        "\"\"\" plot the train and val accuracies \"\"\"\n",
        "# Plot the accuracy\n",
        "utils.plt_metric(history=history.history, metric=\"accuracy\", title=\"Model accuracy\")\n",
        "    \n",
        "# Plot the constrastive loss\n",
        "utils.plt_metric(history=history.history, metric=\"loss\", title=\"Constrastive Loss\")\n",
        "\n",
        "    \n",
        "print(\"End of Training, the model is saved to\", MODEL_FNAME)  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "7JLVdDVI179X",
        "outputId": "917dfd30-e851-497d-dce0-f032146cf8de"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 588 images belonging to 4 classes.\n",
            "\n",
            "The test set contains:\n",
            "• 147 glioma images\n",
            "• 147 meningioma images\n",
            "• 147 normal images\n",
            "• 147 pituitary images\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 150, 150, 3)]     0         \n",
            "                                                                 \n",
            " block1_conv1 (Conv2D)       (None, 150, 150, 64)      1792      \n",
            "                                                                 \n",
            " block1_conv2 (Conv2D)       (None, 150, 150, 64)      36928     \n",
            "                                                                 \n",
            " block1_pool (MaxPooling2D)  (None, 75, 75, 64)        0         \n",
            "                                                                 \n",
            " block2_conv1 (Conv2D)       (None, 75, 75, 128)       73856     \n",
            "                                                                 \n",
            " block2_conv2 (Conv2D)       (None, 75, 75, 128)       147584    \n",
            "                                                                 \n",
            " block2_pool (MaxPooling2D)  (None, 37, 37, 128)       0         \n",
            "                                                                 \n",
            " block3_conv1 (Conv2D)       (None, 37, 37, 256)       295168    \n",
            "                                                                 \n",
            " block3_conv2 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_conv3 (Conv2D)       (None, 37, 37, 256)       590080    \n",
            "                                                                 \n",
            " block3_pool (MaxPooling2D)  (None, 18, 18, 256)       0         \n",
            "                                                                 \n",
            " block4_conv1 (Conv2D)       (None, 18, 18, 512)       1180160   \n",
            "                                                                 \n",
            " block4_conv2 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_conv3 (Conv2D)       (None, 18, 18, 512)       2359808   \n",
            "                                                                 \n",
            " block4_pool (MaxPooling2D)  (None, 9, 9, 512)         0         \n",
            "                                                                 \n",
            " block5_conv1 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv2 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_conv3 (Conv2D)       (None, 9, 9, 512)         2359808   \n",
            "                                                                 \n",
            " block5_pool (MaxPooling2D)  (None, 4, 4, 512)         0         \n",
            "                                                                 \n",
            " global_average_pooling2d (G  (None, 512)              0         \n",
            " lobalAveragePooling2D)                                          \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 512)               0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 512)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 4)                 2052      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 14,716,740\n",
            "Trainable params: 14,716,740\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "\n",
            "Evaluate on test data\n",
            "Accuracy: 0.935374149659864\n",
            "Precision: 0.9367656166253366\n",
            "Recall: 0.935374149659864\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "test_batches = ImageDataGenerator(rescale = 1 / 255.).flow_from_directory(test_dir,\n",
        "                                                        target_size=(INPUT_SIZE,INPUT_SIZE),\n",
        "                                                        class_mode='categorical',\n",
        "                                                        shuffle=False,\n",
        "                                                        seed=42,\n",
        "                                                        batch_size=BATCH_SIZE)\n",
        "    \n",
        "\n",
        "test_glioma_dir = os.path.join(test_dir, 'glioma_tumor')\n",
        "test_meningioma_dir = os.path.join(test_dir, 'meningioma_tumor')\n",
        "test_normal_dir = os.path.join(test_dir, 'no_tumor')\n",
        "test_pituitary_dir = os.path.join(test_dir, 'pituitary_tumor')\n",
        "    \n",
        "num_glioma_test = int(len(os.listdir(test_glioma_dir)))\n",
        "num_meningioma_test = int(len(os.listdir(test_meningioma_dir)))\n",
        "num_normal_test = int(len(os.listdir(test_normal_dir)))\n",
        "num_pituitary_test = int(len(os.listdir(test_pituitary_dir)))\n",
        "\n",
        "print('\\nThe test set contains:')\n",
        "print('\\u2022 {} glioma images'.format(num_glioma_test))\n",
        "print('\\u2022 {} meningioma images'.format(num_meningioma_test))\n",
        "print('\\u2022 {} normal images'.format(num_normal_test))\n",
        "print('\\u2022 {} pituitary images'.format(num_pituitary_test))\n",
        "\n",
        "\"\"\" Test the model \"\"\"\n",
        "model = tf.keras.models.load_model(MODEL_FNAME)\n",
        "model.summary()\n",
        "    \n",
        "y_test = test_batches.classes\n",
        "    \n",
        "#Confution Matrix and Classification Report\n",
        "Y_pred = model.predict_generator(test_batches, (num_glioma_test + num_meningioma_test + num_normal_test + num_pituitary_test) // BATCH_SIZE+1)\n",
        "y_pred = np.argmax(Y_pred, axis=1)\n",
        "    \n",
        "class_labels = list(test_batches.class_indices.keys())   \n",
        "    \n",
        "cm = confusion_matrix(test_batches.classes, y_pred)    \n",
        "cm_display = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=class_labels).plot()\n",
        "    \n",
        "# results = model.evaluate_generator(test_batches)\n",
        "print(\"\\nEvaluate on test data\")\n",
        "print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n",
        "print(\"Precision:\", precision_score(y_test, y_pred, average='weighted'))\n",
        "print(\"Recall:\", recall_score(y_test, y_pred, average='weighted'))\n",
        "    \n",
        "tf.keras.backend.clear_session()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "kljGV_L72ja5"
      },
      "source": [
        "Siamese Network"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rkZn3bx9e35o",
        "outputId": "08c8f6aa-e810-4f5f-e394-ffc7813ebb60"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Using:\n",
            "\t• TensorFlow version: 2.12.0\n",
            "\t• tf.keras version: 2.12.0\n",
            "\t• Running on GPU\n"
          ]
        }
      ],
      "source": [
        "# -*- coding: utf-8 -*-\n",
        "import utils\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import Dense, Input, Flatten, Lambda\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, accuracy_score, precision_score, recall_score, roc_auc_score, f1_score\n",
        "\n",
        "import numpy as np\n",
        "import os\n",
        "\n",
        "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
        "\n",
        "import logging\n",
        "\n",
        "tf.random.set_seed(0)\n",
        "tf.keras.utils.set_random_seed(0)\n",
        "\n",
        "logger = tf.get_logger()\n",
        "logger.setLevel(logging.ERROR)\n",
        "\n",
        "print('Using:')\n",
        "print('\\t\\u2022 TensorFlow version:', tf.__version__)\n",
        "print('\\t\\u2022 tf.keras version:', tf.keras.__version__)\n",
        "print('\\t\\u2022 Running on GPU' if tf.test.is_gpu_available() else '\\t\\u2022 GPU device not found. Running on CPU')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NLPSgD3se-8y",
        "outputId": "228cf209-9999-4fad-dd28-6f13db5bd70c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 40 images belonging to 4 classes.\n",
            "The train set contains 40\n",
            "Found 40 images belonging to 4 classes.\n",
            "The valid set contains 40\n",
            "Found 1920 images belonging to 4 classes.\n",
            "The test set contains 1920\n",
            "number of pairs for training 80\n",
            "number of pairs for validation 80\n",
            "number of pairs for test 3840\n"
          ]
        }
      ],
      "source": [
        "INPUT_SIZE = 150\n",
        "\n",
        "basedir = os.path.join(\"dataset\", \"siamese\") \n",
        "\n",
        "train_image_list, train_y_list = utils.load_images(basedir, 'train', (INPUT_SIZE,INPUT_SIZE))\n",
        "print(\"The train set contains\",len(train_image_list)) \n",
        "\n",
        "valid_image_list, valid_y_list = utils.load_images(basedir, 'validation', (INPUT_SIZE,INPUT_SIZE))   \n",
        "print(\"The valid set contains\", len(valid_image_list))  \n",
        "\n",
        "test_image_list, test_y_list = utils.load_images(basedir, 'test', (INPUT_SIZE,INPUT_SIZE))   \n",
        "print(\"The test set contains\", len(test_image_list))  \n",
        "\n",
        "# make train pairs\n",
        "pairs_train, labels_train = utils.make_pairs(train_image_list, train_y_list)\n",
        "\n",
        "# make validation pairs\n",
        "pairs_val, labels_val = utils.make_pairs(valid_image_list, valid_y_list)\n",
        "\n",
        "# make test pairs\n",
        "pairs_test, labels_test = utils.make_pairs(test_image_list, test_y_list)\n",
        "\n",
        "x_train_1 = pairs_train[:, 0]  \n",
        "x_train_2 = pairs_train[:, 1]\n",
        "print(\"number of pairs for training\", np.shape(x_train_1)[0]) \n",
        "\n",
        "x_val_1 = pairs_val[:, 0] \n",
        "x_val_2 = pairs_val[:, 1]\n",
        "print(\"number of pairs for validation\", np.shape(x_val_1)[0]) \n",
        "\n",
        "x_test_1 = pairs_test[:, 0] \n",
        "x_test_2 = pairs_test[:, 1]\n",
        "print(\"number of pairs for test\", np.shape(x_test_1)[0]) \n",
        "\n",
        "# utils.visualize(pairs_train[:-1], labels_train[:-1], to_show=4, num_col=4)\n",
        "\n",
        "tf.compat.v1.reset_default_graph()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z9iioQiRfDfZ",
        "outputId": "620340de-2074-42b9-bd2a-0ddffc2ada2d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(None, 150, 150, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " input_2 (InputLayer)           [(None, 150, 150, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " sequential (Sequential)        (None, 5120)         14742340    ['input_1[0][0]',                \n",
            "                                                                  'input_2[0][0]']                \n",
            "                                                                                                  \n",
            " lambda (Lambda)                (None, 1)            0           ['sequential[0][0]',             \n",
            "                                                                  'sequential[1][0]']             \n",
            "                                                                                                  \n",
            " dense (Dense)                  (None, 1)            2           ['lambda[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 14,742,342\n",
            "Trainable params: 25,602\n",
            "Non-trainable params: 14,716,740\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "SIAMESE_MODEL_FNAME = 'siamese_network.h5'\n",
        "EMBEDDING_MODEL_FNAME = 'embedding_network.h5'\n",
        "\n",
        "    \n",
        "input_1 = Input((INPUT_SIZE,INPUT_SIZE,3))\n",
        "input_2 = Input((INPUT_SIZE,INPUT_SIZE,3))\n",
        "\n",
        "embedding_network = tf.keras.models.load_model(EMBEDDING_MODEL_FNAME)\n",
        "embedding_network.trainable = False\n",
        "\n",
        "model = tf.keras.Sequential() \n",
        "for layer in embedding_network.layers:  \n",
        "    model.add(layer) \n",
        "\n",
        "model.add(Flatten(name='flat'))\n",
        "model.add(Dense(5120, name='den', activation='sigmoid', kernel_regularizer='l2')) \n",
        " \n",
        "output_1 = model(input_1) \n",
        "output_2 = model(input_2) \n",
        " \n",
        "# merge_layer = Lambda(utils.euclidean_distance)([output_1, output_2]) \n",
        "merge_layer = Lambda(utils.manhattan_distance)([output_1, output_2]) \n",
        "output_layer = Dense(1, activation=\"sigmoid\")(merge_layer) \n",
        "siamese = Model(inputs=[input_1, input_2], outputs=output_layer) \n",
        "siamese.summary()\n",
        "\n",
        "\"\"\" callbacks \"\"\"\n",
        "\n",
        "reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=5, min_lr=0.00001)\n",
        "\n",
        "early_stopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, min_delta=0.0001)\n",
        "\n",
        "checkpointer = ModelCheckpoint(filepath=SIAMESE_MODEL_FNAME, verbose=1, \n",
        "                                save_best_only=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "z2FaS_sof4Dk",
        "outputId": "8655c59c-63d5-4b1b-a868-9bddf1aa7efa"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(None, 150, 150, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " input_2 (InputLayer)           [(None, 150, 150, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " sequential (Sequential)        (None, 5120)         14742340    ['input_1[0][0]',                \n",
            "                                                                  'input_2[0][0]']                \n",
            "                                                                                                  \n",
            " lambda (Lambda)                (None, 1)            0           ['sequential[0][0]',             \n",
            "                                                                  'sequential[1][0]']             \n",
            "                                                                                                  \n",
            " dense (Dense)                  (None, 1)            2           ['lambda[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 14,742,342\n",
            "Trainable params: 25,602\n",
            "Non-trainable params: 14,716,740\n",
            "__________________________________________________________________________________________________\n",
            "Epoch 1/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.2326 - accuracy: 0.7922\n",
            "Epoch 1: val_loss improved from inf to 0.25005, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 6s 39ms/step - loss: 0.2316 - accuracy: 0.8000 - val_loss: 0.2501 - val_accuracy: 0.7875 - lr: 1.0000e-04\n",
            "Epoch 2/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1979 - accuracy: 0.9091\n",
            "Epoch 2: val_loss improved from 0.25005 to 0.22167, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1943 - accuracy: 0.9125 - val_loss: 0.2217 - val_accuracy: 0.7875 - lr: 1.0000e-04\n",
            "Epoch 3/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1779 - accuracy: 0.8974\n",
            "Epoch 3: val_loss improved from 0.22167 to 0.20546, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 33ms/step - loss: 0.1768 - accuracy: 0.9000 - val_loss: 0.2055 - val_accuracy: 0.8250 - lr: 1.0000e-04\n",
            "Epoch 4/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1656 - accuracy: 0.9359\n",
            "Epoch 4: val_loss improved from 0.20546 to 0.18966, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1660 - accuracy: 0.9375 - val_loss: 0.1897 - val_accuracy: 0.8250 - lr: 1.0000e-04\n",
            "Epoch 5/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1639 - accuracy: 0.9359\n",
            "Epoch 5: val_loss improved from 0.18966 to 0.18909, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1664 - accuracy: 0.9250 - val_loss: 0.1891 - val_accuracy: 0.8250 - lr: 1.0000e-04\n",
            "Epoch 6/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1561 - accuracy: 0.9351\n",
            "Epoch 6: val_loss improved from 0.18909 to 0.18674, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 33ms/step - loss: 0.1564 - accuracy: 0.9375 - val_loss: 0.1867 - val_accuracy: 0.8250 - lr: 1.0000e-04\n",
            "Epoch 7/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1500 - accuracy: 0.9351\n",
            "Epoch 7: val_loss improved from 0.18674 to 0.17945, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1504 - accuracy: 0.9375 - val_loss: 0.1795 - val_accuracy: 0.8250 - lr: 1.0000e-04\n",
            "Epoch 8/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1472 - accuracy: 0.9359\n",
            "Epoch 8: val_loss improved from 0.17945 to 0.16595, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1435 - accuracy: 0.9375 - val_loss: 0.1660 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 9/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1424 - accuracy: 0.9487\n",
            "Epoch 9: val_loss improved from 0.16595 to 0.15976, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1537 - accuracy: 0.9375 - val_loss: 0.1598 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 10/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1435 - accuracy: 0.9615\n",
            "Epoch 10: val_loss improved from 0.15976 to 0.15916, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1482 - accuracy: 0.9500 - val_loss: 0.1592 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 11/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1415 - accuracy: 0.9487\n",
            "Epoch 11: val_loss did not improve from 0.15916\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1439 - accuracy: 0.9500 - val_loss: 0.1612 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 12/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1498 - accuracy: 0.9351\n",
            "Epoch 12: val_loss did not improve from 0.15916\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1474 - accuracy: 0.9375 - val_loss: 0.1598 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 13/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1447 - accuracy: 0.9359\n",
            "Epoch 13: val_loss improved from 0.15916 to 0.15721, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1470 - accuracy: 0.9375 - val_loss: 0.1572 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 14/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1415 - accuracy: 0.9103\n",
            "Epoch 14: val_loss did not improve from 0.15721\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1438 - accuracy: 0.9125 - val_loss: 0.1572 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 15/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1466 - accuracy: 0.9351\n",
            "Epoch 15: val_loss did not improve from 0.15721\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1498 - accuracy: 0.9375 - val_loss: 0.1579 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 16/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1366 - accuracy: 0.9487\n",
            "Epoch 16: val_loss improved from 0.15721 to 0.15476, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1389 - accuracy: 0.9500 - val_loss: 0.1548 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 17/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1490 - accuracy: 0.9359\n",
            "Epoch 17: val_loss did not improve from 0.15476\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1482 - accuracy: 0.9375 - val_loss: 0.1548 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 18/175\n",
            "80/80 [==============================] - ETA: 0s - loss: 0.1507 - accuracy: 0.9250\n",
            "Epoch 18: val_loss did not improve from 0.15476\n",
            "80/80 [==============================] - 3s 33ms/step - loss: 0.1507 - accuracy: 0.9250 - val_loss: 0.1560 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 19/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1348 - accuracy: 0.9351\n",
            "Epoch 19: val_loss improved from 0.15476 to 0.15423, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1387 - accuracy: 0.9250 - val_loss: 0.1542 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 20/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1442 - accuracy: 0.9359\n",
            "Epoch 20: val_loss did not improve from 0.15423\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1526 - accuracy: 0.9250 - val_loss: 0.1568 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 21/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1435 - accuracy: 0.9351\n",
            "Epoch 21: val_loss improved from 0.15423 to 0.15275, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1466 - accuracy: 0.9375 - val_loss: 0.1528 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 22/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1410 - accuracy: 0.9487\n",
            "Epoch 22: val_loss improved from 0.15275 to 0.15065, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1403 - accuracy: 0.9500 - val_loss: 0.1506 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 23/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1476 - accuracy: 0.9221\n",
            "Epoch 23: val_loss did not improve from 0.15065\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1449 - accuracy: 0.9250 - val_loss: 0.1545 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 24/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1532 - accuracy: 0.9487\n",
            "Epoch 24: val_loss did not improve from 0.15065\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1549 - accuracy: 0.9500 - val_loss: 0.1516 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 25/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1412 - accuracy: 0.9359\n",
            "Epoch 25: val_loss improved from 0.15065 to 0.14861, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1378 - accuracy: 0.9375 - val_loss: 0.1486 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 26/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1479 - accuracy: 0.9487\n",
            "Epoch 26: val_loss did not improve from 0.14861\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1442 - accuracy: 0.9500 - val_loss: 0.1538 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 27/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1466 - accuracy: 0.9351\n",
            "Epoch 27: val_loss did not improve from 0.14861\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1439 - accuracy: 0.9375 - val_loss: 0.1526 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 28/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1408 - accuracy: 0.9487\n",
            "Epoch 28: val_loss did not improve from 0.14861\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1400 - accuracy: 0.9500 - val_loss: 0.1493 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 29/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1386 - accuracy: 0.9221\n",
            "Epoch 29: val_loss improved from 0.14861 to 0.14695, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1388 - accuracy: 0.9250 - val_loss: 0.1470 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 30/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1394 - accuracy: 0.9231\n",
            "Epoch 30: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1386 - accuracy: 0.9250 - val_loss: 0.1474 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 31/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1407 - accuracy: 0.9103\n",
            "Epoch 31: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1399 - accuracy: 0.9125 - val_loss: 0.1531 - val_accuracy: 0.8625 - lr: 1.0000e-04\n",
            "Epoch 32/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1269 - accuracy: 0.9351\n",
            "Epoch 32: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1275 - accuracy: 0.9375 - val_loss: 0.1495 - val_accuracy: 0.9000 - lr: 1.0000e-04\n",
            "Epoch 33/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1437 - accuracy: 0.9359\n",
            "Epoch 33: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1427 - accuracy: 0.9375 - val_loss: 0.1478 - val_accuracy: 0.9000 - lr: 1.0000e-04\n",
            "Epoch 34/175\n",
            "79/80 [============================>.] - ETA: 0s - loss: 0.1375 - accuracy: 0.9114\n",
            "Epoch 34: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1358 - accuracy: 0.9125 - val_loss: 0.1483 - val_accuracy: 0.9000 - lr: 1.0000e-04\n",
            "Epoch 35/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1537 - accuracy: 0.9103\n",
            "Epoch 35: val_loss did not improve from 0.14695\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1499 - accuracy: 0.9125 - val_loss: 0.1473 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 36/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1468 - accuracy: 0.9231\n",
            "Epoch 36: val_loss improved from 0.14695 to 0.14619, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1432 - accuracy: 0.9250 - val_loss: 0.1462 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 37/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1325 - accuracy: 0.9487\n",
            "Epoch 37: val_loss improved from 0.14619 to 0.14592, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1326 - accuracy: 0.9375 - val_loss: 0.1459 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 38/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1334 - accuracy: 0.9487\n",
            "Epoch 38: val_loss improved from 0.14592 to 0.14579, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 31ms/step - loss: 0.1339 - accuracy: 0.9375 - val_loss: 0.1458 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 39/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1332 - accuracy: 0.9091\n",
            "Epoch 39: val_loss improved from 0.14579 to 0.14560, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1308 - accuracy: 0.9125 - val_loss: 0.1456 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 40/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1372 - accuracy: 0.9221\n",
            "Epoch 40: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1360 - accuracy: 0.9125 - val_loss: 0.1468 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 41/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1315 - accuracy: 0.9091\n",
            "Epoch 41: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1321 - accuracy: 0.9125 - val_loss: 0.1471 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 42/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1336 - accuracy: 0.9359\n",
            "Epoch 42: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1303 - accuracy: 0.9375 - val_loss: 0.1471 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 43/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1425 - accuracy: 0.9231\n",
            "Epoch 43: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1415 - accuracy: 0.9250 - val_loss: 0.1483 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 44/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1316 - accuracy: 0.9221\n",
            "Epoch 44: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1294 - accuracy: 0.9250 - val_loss: 0.1484 - val_accuracy: 0.9000 - lr: 2.0000e-05\n",
            "Epoch 45/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1304 - accuracy: 0.9481\n",
            "Epoch 45: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1282 - accuracy: 0.9500 - val_loss: 0.1472 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 46/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1321 - accuracy: 0.9103\n",
            "Epoch 46: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1320 - accuracy: 0.9125 - val_loss: 0.1477 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 47/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1299 - accuracy: 0.9351\n",
            "Epoch 47: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1303 - accuracy: 0.9375 - val_loss: 0.1487 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 48/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1253 - accuracy: 0.9221\n",
            "Epoch 48: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1267 - accuracy: 0.9250 - val_loss: 0.1485 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 49/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1269 - accuracy: 0.9359\n",
            "Epoch 49: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1273 - accuracy: 0.9250 - val_loss: 0.1485 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 50/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1323 - accuracy: 0.9231\n",
            "Epoch 50: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1342 - accuracy: 0.9250 - val_loss: 0.1481 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 51/175\n",
            "79/80 [============================>.] - ETA: 0s - loss: 0.1263 - accuracy: 0.9241\n",
            "Epoch 51: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1318 - accuracy: 0.9125 - val_loss: 0.1480 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 52/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1362 - accuracy: 0.9231\n",
            "Epoch 52: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1354 - accuracy: 0.9250 - val_loss: 0.1496 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 53/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1323 - accuracy: 0.9359\n",
            "Epoch 53: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1291 - accuracy: 0.9375 - val_loss: 0.1501 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 54/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1205 - accuracy: 0.9351\n",
            "Epoch 54: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1278 - accuracy: 0.9250 - val_loss: 0.1480 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 55/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1314 - accuracy: 0.9103\n",
            "Epoch 55: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1308 - accuracy: 0.9125 - val_loss: 0.1504 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 56/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1260 - accuracy: 0.9487\n",
            "Epoch 56: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1300 - accuracy: 0.9375 - val_loss: 0.1477 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 57/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1260 - accuracy: 0.9231\n",
            "Epoch 57: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1264 - accuracy: 0.9250 - val_loss: 0.1492 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 58/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1317 - accuracy: 0.9351\n",
            "Epoch 58: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1269 - accuracy: 0.9375 - val_loss: 0.1474 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 59/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1279 - accuracy: 0.9351\n",
            "Epoch 59: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1257 - accuracy: 0.9375 - val_loss: 0.1475 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 60/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1315 - accuracy: 0.9221\n",
            "Epoch 60: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1343 - accuracy: 0.9250 - val_loss: 0.1477 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 61/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1236 - accuracy: 0.9481\n",
            "Epoch 61: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1323 - accuracy: 0.9375 - val_loss: 0.1487 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 62/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1308 - accuracy: 0.9231\n",
            "Epoch 62: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1276 - accuracy: 0.9250 - val_loss: 0.1484 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 63/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1243 - accuracy: 0.9610\n",
            "Epoch 63: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1332 - accuracy: 0.9375 - val_loss: 0.1460 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 64/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1315 - accuracy: 0.9221\n",
            "Epoch 64: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1318 - accuracy: 0.9250 - val_loss: 0.1479 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 65/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1336 - accuracy: 0.9231\n",
            "Epoch 65: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1328 - accuracy: 0.9250 - val_loss: 0.1480 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 66/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1348 - accuracy: 0.9221\n",
            "Epoch 66: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1346 - accuracy: 0.9250 - val_loss: 0.1494 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 67/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1291 - accuracy: 0.9221\n",
            "Epoch 67: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1269 - accuracy: 0.9250 - val_loss: 0.1482 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 68/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1360 - accuracy: 0.9221\n",
            "Epoch 68: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1360 - accuracy: 0.9250 - val_loss: 0.1487 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 69/175\n",
            "79/80 [============================>.] - ETA: 0s - loss: 0.1238 - accuracy: 0.9494\n",
            "Epoch 69: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1248 - accuracy: 0.9500 - val_loss: 0.1482 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 70/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1207 - accuracy: 0.9615\n",
            "Epoch 70: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1312 - accuracy: 0.9375 - val_loss: 0.1478 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 71/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1299 - accuracy: 0.8974\n",
            "Epoch 71: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1302 - accuracy: 0.9000 - val_loss: 0.1486 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 72/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1347 - accuracy: 0.9231\n",
            "Epoch 72: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1314 - accuracy: 0.9250 - val_loss: 0.1481 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 73/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1302 - accuracy: 0.9221\n",
            "Epoch 73: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1304 - accuracy: 0.9250 - val_loss: 0.1490 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 74/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1289 - accuracy: 0.9231\n",
            "Epoch 74: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1267 - accuracy: 0.9250 - val_loss: 0.1488 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 75/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1304 - accuracy: 0.9359\n",
            "Epoch 75: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1281 - accuracy: 0.9375 - val_loss: 0.1479 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 76/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1250 - accuracy: 0.9487\n",
            "Epoch 76: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1276 - accuracy: 0.9375 - val_loss: 0.1478 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 77/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1337 - accuracy: 0.9091\n",
            "Epoch 77: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1313 - accuracy: 0.9125 - val_loss: 0.1482 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 78/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1206 - accuracy: 0.9481\n",
            "Epoch 78: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1219 - accuracy: 0.9500 - val_loss: 0.1468 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 79/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1269 - accuracy: 0.9359\n",
            "Epoch 79: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1297 - accuracy: 0.9250 - val_loss: 0.1470 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 80/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1336 - accuracy: 0.9231\n",
            "Epoch 80: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1328 - accuracy: 0.9250 - val_loss: 0.1469 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 81/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1348 - accuracy: 0.9351\n",
            "Epoch 81: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1323 - accuracy: 0.9375 - val_loss: 0.1474 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 82/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1289 - accuracy: 0.9221\n",
            "Epoch 82: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1292 - accuracy: 0.9250 - val_loss: 0.1483 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 83/175\n",
            "79/80 [============================>.] - ETA: 0s - loss: 0.1309 - accuracy: 0.9241\n",
            "Epoch 83: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1303 - accuracy: 0.9250 - val_loss: 0.1482 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 84/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1339 - accuracy: 0.9221\n",
            "Epoch 84: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1300 - accuracy: 0.9250 - val_loss: 0.1487 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 85/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1300 - accuracy: 0.9487\n",
            "Epoch 85: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1267 - accuracy: 0.9500 - val_loss: 0.1468 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 86/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1312 - accuracy: 0.9103\n",
            "Epoch 86: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1287 - accuracy: 0.9125 - val_loss: 0.1462 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 87/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1353 - accuracy: 0.9103\n",
            "Epoch 87: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1345 - accuracy: 0.9125 - val_loss: 0.1476 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 88/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1374 - accuracy: 0.9091\n",
            "Epoch 88: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1323 - accuracy: 0.9125 - val_loss: 0.1474 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 89/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1306 - accuracy: 0.9231\n",
            "Epoch 89: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1299 - accuracy: 0.9250 - val_loss: 0.1480 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 90/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1341 - accuracy: 0.9091\n",
            "Epoch 90: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1335 - accuracy: 0.9125 - val_loss: 0.1483 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 91/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1249 - accuracy: 0.9359\n",
            "Epoch 91: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1243 - accuracy: 0.9375 - val_loss: 0.1475 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 92/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1318 - accuracy: 0.9351\n",
            "Epoch 92: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1294 - accuracy: 0.9375 - val_loss: 0.1466 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 93/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1323 - accuracy: 0.9221\n",
            "Epoch 93: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1300 - accuracy: 0.9250 - val_loss: 0.1484 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 94/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1299 - accuracy: 0.9231\n",
            "Epoch 94: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1270 - accuracy: 0.9250 - val_loss: 0.1470 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 95/175\n",
            "80/80 [==============================] - ETA: 0s - loss: 0.1262 - accuracy: 0.9125\n",
            "Epoch 95: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1262 - accuracy: 0.9125 - val_loss: 0.1472 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 96/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1252 - accuracy: 0.9221\n",
            "Epoch 96: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1257 - accuracy: 0.9250 - val_loss: 0.1470 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 97/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1264 - accuracy: 0.9359\n",
            "Epoch 97: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1265 - accuracy: 0.9375 - val_loss: 0.1476 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 98/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1285 - accuracy: 0.9231\n",
            "Epoch 98: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1304 - accuracy: 0.9250 - val_loss: 0.1473 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 99/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1294 - accuracy: 0.8974\n",
            "Epoch 99: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1262 - accuracy: 0.9000 - val_loss: 0.1475 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 100/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1341 - accuracy: 0.9221\n",
            "Epoch 100: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1361 - accuracy: 0.9125 - val_loss: 0.1483 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 101/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1263 - accuracy: 0.9103\n",
            "Epoch 101: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1257 - accuracy: 0.9125 - val_loss: 0.1486 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 102/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1277 - accuracy: 0.9351\n",
            "Epoch 102: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1254 - accuracy: 0.9375 - val_loss: 0.1457 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 103/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1305 - accuracy: 0.9221\n",
            "Epoch 103: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1282 - accuracy: 0.9250 - val_loss: 0.1469 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 104/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1202 - accuracy: 0.9481\n",
            "Epoch 104: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1190 - accuracy: 0.9375 - val_loss: 0.1460 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 105/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1298 - accuracy: 0.9231\n",
            "Epoch 105: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1266 - accuracy: 0.9250 - val_loss: 0.1472 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 106/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1319 - accuracy: 0.9231\n",
            "Epoch 106: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1312 - accuracy: 0.9250 - val_loss: 0.1485 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 107/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1260 - accuracy: 0.9481\n",
            "Epoch 107: val_loss did not improve from 0.14560\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1238 - accuracy: 0.9500 - val_loss: 0.1466 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 108/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1280 - accuracy: 0.9221\n",
            "Epoch 108: val_loss improved from 0.14560 to 0.14530, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1258 - accuracy: 0.9250 - val_loss: 0.1453 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 109/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1272 - accuracy: 0.9359\n",
            "Epoch 109: val_loss did not improve from 0.14530\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1241 - accuracy: 0.9375 - val_loss: 0.1454 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 110/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1315 - accuracy: 0.9231\n",
            "Epoch 110: val_loss did not improve from 0.14530\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1315 - accuracy: 0.9250 - val_loss: 0.1461 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 111/175\n",
            "79/80 [============================>.] - ETA: 0s - loss: 0.1149 - accuracy: 0.9620\n",
            "Epoch 111: val_loss improved from 0.14530 to 0.14416, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1162 - accuracy: 0.9625 - val_loss: 0.1442 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 112/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1383 - accuracy: 0.8974\n",
            "Epoch 112: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1374 - accuracy: 0.9000 - val_loss: 0.1448 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 113/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1300 - accuracy: 0.9351\n",
            "Epoch 113: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1312 - accuracy: 0.9250 - val_loss: 0.1456 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 114/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1276 - accuracy: 0.9221\n",
            "Epoch 114: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1267 - accuracy: 0.9250 - val_loss: 0.1464 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 115/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1290 - accuracy: 0.9615\n",
            "Epoch 115: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1258 - accuracy: 0.9625 - val_loss: 0.1444 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 116/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1224 - accuracy: 0.9359\n",
            "Epoch 116: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1229 - accuracy: 0.9250 - val_loss: 0.1445 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 117/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1254 - accuracy: 0.9481\n",
            "Epoch 117: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1233 - accuracy: 0.9500 - val_loss: 0.1452 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 118/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1218 - accuracy: 0.9359\n",
            "Epoch 118: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1245 - accuracy: 0.9250 - val_loss: 0.1445 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 119/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1251 - accuracy: 0.9221\n",
            "Epoch 119: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1230 - accuracy: 0.9250 - val_loss: 0.1448 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 120/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1222 - accuracy: 0.9481\n",
            "Epoch 120: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1237 - accuracy: 0.9500 - val_loss: 0.1450 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 121/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1334 - accuracy: 0.9231\n",
            "Epoch 121: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1328 - accuracy: 0.9250 - val_loss: 0.1446 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 122/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1212 - accuracy: 0.9487\n",
            "Epoch 122: val_loss did not improve from 0.14416\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1207 - accuracy: 0.9500 - val_loss: 0.1446 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 123/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1148 - accuracy: 0.9744\n",
            "Epoch 123: val_loss improved from 0.14416 to 0.14339, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1278 - accuracy: 0.9500 - val_loss: 0.1434 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 124/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1310 - accuracy: 0.9231\n",
            "Epoch 124: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1302 - accuracy: 0.9250 - val_loss: 0.1454 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 125/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1159 - accuracy: 0.9744\n",
            "Epoch 125: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1193 - accuracy: 0.9625 - val_loss: 0.1442 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 126/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1236 - accuracy: 0.9221\n",
            "Epoch 126: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1265 - accuracy: 0.9250 - val_loss: 0.1443 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 127/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1181 - accuracy: 0.9487\n",
            "Epoch 127: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1201 - accuracy: 0.9500 - val_loss: 0.1435 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 128/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1299 - accuracy: 0.9091\n",
            "Epoch 128: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1301 - accuracy: 0.9125 - val_loss: 0.1440 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 129/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1301 - accuracy: 0.9487\n",
            "Epoch 129: val_loss did not improve from 0.14339\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1268 - accuracy: 0.9500 - val_loss: 0.1436 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 130/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1191 - accuracy: 0.9481\n",
            "Epoch 130: val_loss improved from 0.14339 to 0.14250, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1196 - accuracy: 0.9500 - val_loss: 0.1425 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 131/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1269 - accuracy: 0.9351\n",
            "Epoch 131: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1296 - accuracy: 0.9375 - val_loss: 0.1436 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 132/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1195 - accuracy: 0.9615\n",
            "Epoch 132: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1215 - accuracy: 0.9625 - val_loss: 0.1437 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 133/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1308 - accuracy: 0.9221\n",
            "Epoch 133: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1310 - accuracy: 0.9250 - val_loss: 0.1439 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 134/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1229 - accuracy: 0.9359\n",
            "Epoch 134: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1234 - accuracy: 0.9250 - val_loss: 0.1436 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 135/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1231 - accuracy: 0.9231\n",
            "Epoch 135: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1250 - accuracy: 0.9250 - val_loss: 0.1437 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 136/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1321 - accuracy: 0.9091\n",
            "Epoch 136: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1297 - accuracy: 0.9125 - val_loss: 0.1446 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 137/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1257 - accuracy: 0.9351\n",
            "Epoch 137: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1260 - accuracy: 0.9375 - val_loss: 0.1448 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 138/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1257 - accuracy: 0.9221\n",
            "Epoch 138: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1259 - accuracy: 0.9250 - val_loss: 0.1447 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 139/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1244 - accuracy: 0.9091\n",
            "Epoch 139: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1247 - accuracy: 0.9125 - val_loss: 0.1445 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 140/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1271 - accuracy: 0.9351\n",
            "Epoch 140: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1248 - accuracy: 0.9375 - val_loss: 0.1443 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 141/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1208 - accuracy: 0.9487\n",
            "Epoch 141: val_loss did not improve from 0.14250\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1203 - accuracy: 0.9500 - val_loss: 0.1429 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 142/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1178 - accuracy: 0.9351\n",
            "Epoch 142: val_loss improved from 0.14250 to 0.14241, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 28ms/step - loss: 0.1184 - accuracy: 0.9375 - val_loss: 0.1424 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 143/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1176 - accuracy: 0.9359\n",
            "Epoch 143: val_loss improved from 0.14241 to 0.14238, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1196 - accuracy: 0.9375 - val_loss: 0.1424 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 144/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1290 - accuracy: 0.9231\n",
            "Epoch 144: val_loss improved from 0.14238 to 0.14226, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1258 - accuracy: 0.9250 - val_loss: 0.1423 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 145/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1278 - accuracy: 0.9351\n",
            "Epoch 145: val_loss improved from 0.14226 to 0.14220, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1280 - accuracy: 0.9375 - val_loss: 0.1422 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 146/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1262 - accuracy: 0.9221\n",
            "Epoch 146: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1239 - accuracy: 0.9250 - val_loss: 0.1423 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 147/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1230 - accuracy: 0.9351\n",
            "Epoch 147: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1209 - accuracy: 0.9375 - val_loss: 0.1427 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 148/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1230 - accuracy: 0.9615\n",
            "Epoch 148: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1283 - accuracy: 0.9500 - val_loss: 0.1422 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 149/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1315 - accuracy: 0.9103\n",
            "Epoch 149: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1332 - accuracy: 0.9125 - val_loss: 0.1437 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 150/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1188 - accuracy: 0.9351\n",
            "Epoch 150: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1234 - accuracy: 0.9250 - val_loss: 0.1438 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 151/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1314 - accuracy: 0.9221\n",
            "Epoch 151: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1289 - accuracy: 0.9250 - val_loss: 0.1462 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 152/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1259 - accuracy: 0.9231\n",
            "Epoch 152: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1253 - accuracy: 0.9250 - val_loss: 0.1440 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 153/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1305 - accuracy: 0.9359\n",
            "Epoch 153: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1297 - accuracy: 0.9375 - val_loss: 0.1427 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 154/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1277 - accuracy: 0.9221\n",
            "Epoch 154: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1279 - accuracy: 0.9250 - val_loss: 0.1431 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 155/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1272 - accuracy: 0.9359\n",
            "Epoch 155: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1265 - accuracy: 0.9375 - val_loss: 0.1424 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 156/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1249 - accuracy: 0.9221\n",
            "Epoch 156: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1228 - accuracy: 0.9250 - val_loss: 0.1425 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 157/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1335 - accuracy: 0.9103\n",
            "Epoch 157: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1302 - accuracy: 0.9125 - val_loss: 0.1441 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 158/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1232 - accuracy: 0.9359\n",
            "Epoch 158: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1236 - accuracy: 0.9375 - val_loss: 0.1443 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 159/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1236 - accuracy: 0.9359\n",
            "Epoch 159: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1231 - accuracy: 0.9375 - val_loss: 0.1436 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 160/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1227 - accuracy: 0.9221\n",
            "Epoch 160: val_loss did not improve from 0.14220\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1248 - accuracy: 0.9125 - val_loss: 0.1436 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 161/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1293 - accuracy: 0.9351\n",
            "Epoch 161: val_loss improved from 0.14220 to 0.14181, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 2s 29ms/step - loss: 0.1294 - accuracy: 0.9375 - val_loss: 0.1418 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 162/175\n",
            "80/80 [==============================] - ETA: 0s - loss: 0.1265 - accuracy: 0.9250\n",
            "Epoch 162: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1265 - accuracy: 0.9250 - val_loss: 0.1426 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 163/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1314 - accuracy: 0.9351\n",
            "Epoch 163: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1297 - accuracy: 0.9375 - val_loss: 0.1429 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 164/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1299 - accuracy: 0.9103\n",
            "Epoch 164: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1291 - accuracy: 0.9125 - val_loss: 0.1434 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 165/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1224 - accuracy: 0.9615\n",
            "Epoch 165: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1242 - accuracy: 0.9625 - val_loss: 0.1428 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 166/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1297 - accuracy: 0.9103\n",
            "Epoch 166: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1290 - accuracy: 0.9125 - val_loss: 0.1444 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 167/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1238 - accuracy: 0.9231\n",
            "Epoch 167: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1231 - accuracy: 0.9250 - val_loss: 0.1440 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 168/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1210 - accuracy: 0.9351\n",
            "Epoch 168: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1213 - accuracy: 0.9375 - val_loss: 0.1430 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 169/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1269 - accuracy: 0.9221\n",
            "Epoch 169: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 30ms/step - loss: 0.1270 - accuracy: 0.9250 - val_loss: 0.1428 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 170/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1200 - accuracy: 0.9231\n",
            "Epoch 170: val_loss did not improve from 0.14181\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1218 - accuracy: 0.9250 - val_loss: 0.1430 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 171/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1145 - accuracy: 0.9610\n",
            "Epoch 171: val_loss improved from 0.14181 to 0.14170, saving model to siamese_network.h5\n",
            "80/80 [==============================] - 3s 32ms/step - loss: 0.1230 - accuracy: 0.9500 - val_loss: 0.1417 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 172/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1130 - accuracy: 0.9610\n",
            "Epoch 172: val_loss did not improve from 0.14170\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1171 - accuracy: 0.9500 - val_loss: 0.1421 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 173/175\n",
            "78/80 [============================>.] - ETA: 0s - loss: 0.1236 - accuracy: 0.9231\n",
            "Epoch 173: val_loss did not improve from 0.14170\n",
            "80/80 [==============================] - 2s 26ms/step - loss: 0.1230 - accuracy: 0.9250 - val_loss: 0.1421 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 174/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1237 - accuracy: 0.9221\n",
            "Epoch 174: val_loss did not improve from 0.14170\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1239 - accuracy: 0.9250 - val_loss: 0.1428 - val_accuracy: 0.9000 - lr: 1.0000e-05\n",
            "Epoch 175/175\n",
            "77/80 [===========================>..] - ETA: 0s - loss: 0.1238 - accuracy: 0.9351\n",
            "Epoch 175: val_loss did not improve from 0.14170\n",
            "80/80 [==============================] - 2s 27ms/step - loss: 0.1226 - accuracy: 0.9375 - val_loss: 0.1440 - val_accuracy: 0.9000 - lr: 1.0000e-05\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\"\"\" train the model \"\"\"\n",
        "\n",
        "optimizer = Adam(learning_rate=0.0001)\n",
        "siamese.compile(loss=utils.loss(1), optimizer=optimizer, metrics=[\"accuracy\"]) # Manhattan [loss: 0.0997 - accuracy: 0.9539] | Euclidean [loss: 0.2267 - accuracy: 0.9487]\n",
        "# siamese.compile(loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), optimizer=optimizer, metrics=[\"accuracy\"]) # Manhattan [loss: 0.4305 - accuracy: 0.9430] | Euclidean [loss: 0.6142 - accuracy: 0.9464]\n",
        "\n",
        "siamese.summary()\n",
        "history = siamese.fit([x_train_1, x_train_2],\n",
        "    labels_train,\n",
        "    validation_data=([x_val_1, x_val_2], labels_val),\n",
        "    batch_size=1,\n",
        "    epochs=175,   # 175 for contrastive 100 for cross ent\n",
        "    callbacks = [checkpointer, reduce_lr]#, early_stopping]\n",
        ")\n",
        "\n",
        "# Plot the accuracy\n",
        "utils.plt_metric(history=history.history, metric=\"accuracy\", title=\"Model accuracy\")\n",
        "\n",
        "# Plot the constrastive loss\n",
        "utils.plt_metric(history=history.history, metric=\"loss\", title=\"Constrastive Loss\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 640
        },
        "id": "1Ya59cg6gFVJ",
        "outputId": "b2869d8b-23db-4c94-f97a-6785d8fdd111"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "120/120 [==============================] - 26s 123ms/step - loss: 0.1279 - accuracy: 0.9453\n",
            "test loss, test acc: [0.12794193625450134, 0.9453125]\n",
            "120/120 [==============================] - 15s 121ms/step\n",
            "\n",
            "Evaluate on test data\n",
            "Accuracy: 0.9453125\n",
            "Precision: 0.9478315524827152\n",
            "Recall: 0.9453125\n",
            "ROC AUC: 0.9453124999999999\n",
            "F1: 0.945235487404162\n",
            "Specificity: 0.9828125\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\"\"\" Test the model \"\"\"\n",
        "siamese = tf.keras.models.load_model(SIAMESE_MODEL_FNAME, custom_objects={'contrastive_loss': utils.loss(1)})\n",
        "\n",
        "results = siamese.evaluate([x_test_1, x_test_2], labels_test)\n",
        "print(\"test loss, test acc:\", results)\n",
        "\n",
        "Y_pred = siamese.predict([x_test_1, x_test_2]).squeeze()\n",
        "y_pred = Y_pred > 0.5\n",
        "y_test = labels_test\n",
        "\n",
        "print(\"\\nEvaluate on test data\")\n",
        "print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n",
        "print(\"Precision:\", precision_score(y_test, y_pred, average='weighted'))\n",
        "print(\"Recall:\", recall_score(y_test, y_pred, average='weighted'))\n",
        "print(\"ROC AUC:\", roc_auc_score(y_test, y_pred, average='weighted'))\n",
        "print(\"F1:\", f1_score(y_test, y_pred, average='weighted'))\n",
        "\n",
        "cm = confusion_matrix(y_test, y_pred)    \n",
        "cm_display = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=[0,1]).plot()\n",
        "\n",
        "tn, fp, fn, tp = cm.ravel()\n",
        "specificity = tn / (tn+fp)\n",
        "print(\"Specificity:\", specificity)\n",
        "\n",
        "tf.keras.backend.clear_session()"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}