2D_CNN+LSTM.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/alexsalman/CSE247/blob/main/2D_CNN%2BLSTM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fk8TLrPiv-ab"
      },
      "source": [
        "####**Convolutional Neural Network + Long Short Term Memory**\n",
        "######*I am using a Convolution Neural Network (CNN) + Long Short Term Memory (LSTM) Network to extract general representation while utilizing the Spatial-temporal aspect of the videos.*"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "M8ibtd5HKtZk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "387e8079-1f6b-499b-fb34-0d8eced81b58"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.8.0\n"
          ]
        }
      ],
      "source": [
        "# required libraries\n",
        "import os\n",
        "import cv2\n",
        "import random\n",
        "import numpy as np\n",
        "import datetime as dt\n",
        "import matplotlib.pyplot as plt\n",
        "from google.colab import drive\n",
        "from sklearn.model_selection import train_test_split\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D, Flatten, Dense, Dropout, Activation, BatchNormalization\n",
        "from keras.layers import TimeDistributed, MaxPooling2D, GlobalAveragePooling2D\n",
        "from keras.layers.recurrent import LSTM\n",
        "from keras import regularizers\n",
        "%matplotlib inline\n",
        "print(tf.version.VERSION)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "iffdFOf1CEAN"
      },
      "outputs": [],
      "source": [
        "# set Numpy, Python, and Tensorflow seeds to get consistent results on every execution\n",
        "seed_constant = 27\n",
        "np.random.seed(seed_constant)\n",
        "random.seed(seed_constant)\n",
        "tf.random.set_seed(seed_constant)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "mcLh22LiOHyn",
        "outputId": "cb71c858-cf56-4ff4-e281-d8ccf4371adc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/gdrive/\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'/content/gdrive/My Drive/247'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 3
        }
      ],
      "source": [
        "# mount dataset from google drive\n",
        "drive.mount('/content/gdrive/', force_remount=True)\n",
        "gdrive_path = '/content/gdrive' + '/My Drive/247/'\n",
        "os.chdir(gdrive_path)\n",
        "os.getcwd()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "oeDK8SzumZ1Q"
      },
      "outputs": [],
      "source": [
        "# frame dimention\n",
        "IMAGE_HEIGHT, IMAGE_WIDTH = 128, 128\n",
        "# frame number for each video (depth)\n",
        "SEQUENCE_LENGTH = 16\n",
        "# video dir path\n",
        "DATASET_DIR = gdrive_path + 'Cropped_videos'\n",
        "# labels of classes\n",
        "CLASSES_LIST = ['hemostasis', 'inflammatory', 'proliferative', 'maturation']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "3mFB5qD6b3Kd"
      },
      "outputs": [],
      "source": [
        "# image cropping\n",
        "def crop_center_square(frame):\n",
        "    y, x = frame.shape[0:2]\n",
        "    min_dim = min(y, x)\n",
        "    start_x = (x // 2) - (min_dim // 2)\n",
        "    start_y = (y // 2) - (min_dim // 2)\n",
        "    return frame[start_y:start_y+min_dim,start_x:start_x+min_dim]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "sqeexMUjaLzJ"
      },
      "outputs": [],
      "source": [
        "def load_video(path, resize=(128, 128)):\n",
        "    video_reader = cv2.VideoCapture(path)\n",
        "    frames = []\n",
        "    try:\n",
        "        while True:\n",
        "            ret, frame = video_reader.read()\n",
        "            if not ret:\n",
        "                  break\n",
        "            frame = crop_center_square(frame)\n",
        "            frame = cv2.resize(frame, resize)\n",
        "            frame = frame[:, :, [2, 1, 0]]\n",
        "            frames.append(frame)\n",
        "    finally:\n",
        "        video_reader.release()\n",
        "    return np.array(frames) / 255.0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "ljUWHW6Jqzu-"
      },
      "outputs": [],
      "source": [
        "def create_dataset(state):\n",
        "    # Declared Empty Lists to store the features, labels and video file path values.\n",
        "    features = []\n",
        "    labels = []\n",
        "    video_files_paths = []\n",
        "    # Iterating through all the classes mentioned in the classes list\n",
        "    for class_index, class_name in enumerate(CLASSES_LIST):\n",
        "        # Display the name of the class whose data is being extracted.\n",
        "        print(f'Extracting Data of Class: {class_name} {state}')\n",
        "        # Get the list of video files present in the specific class name directory.\n",
        "        files_list = os.listdir(os.path.join(DATASET_DIR, class_name))\n",
        "        # Iterate through all the files present in the files list.\n",
        "        for file_name in files_list:\n",
        "            # Get the complete video path.\n",
        "            video_file_path = os.path.join(DATASET_DIR, class_name, file_name)\n",
        "            # create testing data\n",
        "            if state == 'test':\n",
        "                # get the mouse number\n",
        "                mouse_number = int(video_file_path.split(' ')[2].split('_')[1].split('-')[1])\n",
        "                # get the mouse side (L or R)\n",
        "                mouse_side = video_file_path.split(' ')[2].split('_')[1].split('-')[2]\n",
        "                if mouse_number == 4 and mouse_side == 'L':\n",
        "                    frames = load_video(video_file_path)\n",
        "                    features.append(frames)\n",
        "                    labels.append(class_index)\n",
        "                    video_files_paths.append(video_file_path)\n",
        "            # create validation data\n",
        "            elif state == 'valid':\n",
        "                # get the mouse number\n",
        "                mouse_number = int(video_file_path.split(' ')[2].split('_')[1].split('-')[1])\n",
        "                # get the mouse side (L or R)\n",
        "                mouse_side = video_file_path.split(' ')[2].split('_')[1].split('-')[2]\n",
        "                if mouse_number == 4 and mouse_side == 'R':\n",
        "                    frames = load_video(video_file_path)\n",
        "                    features.append(frames)\n",
        "                    labels.append(class_index)\n",
        "                    video_files_paths.append(video_file_path)\n",
        "            # create training data\n",
        "            else:\n",
        "                # get the mouse number\n",
        "                mouse_number = int(video_file_path.split(' ')[2].split('_')[1].split('-')[1])\n",
        "                if mouse_number != 4:\n",
        "                    frames = load_video(video_file_path)\n",
        "                    features.append(frames)\n",
        "                    labels.append(class_index)\n",
        "                    video_files_paths.append(video_file_path)\n",
        "    # Converting the list to numpy arrays\n",
        "    features = np.asarray(features)\n",
        "    # print(features)\n",
        "    labels = np.array(labels)\n",
        "    # Return the frames, class index, and video file path.\n",
        "    return features, labels, video_files_paths"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a8rpanz9rASe",
        "outputId": "432622ba-3670-4750-cb6f-23ed1a08a89f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting Data of Class: hemostasis train\n",
            "Extracting Data of Class: inflammatory train\n",
            "Extracting Data of Class: proliferative train\n",
            "Extracting Data of Class: maturation train\n",
            "Extracting Data of Class: hemostasis test\n",
            "Extracting Data of Class: inflammatory test\n",
            "Extracting Data of Class: proliferative test\n",
            "Extracting Data of Class: maturation test\n",
            "Extracting Data of Class: hemostasis valid\n",
            "Extracting Data of Class: inflammatory valid\n",
            "Extracting Data of Class: proliferative valid\n",
            "Extracting Data of Class: maturation valid\n"
          ]
        }
      ],
      "source": [
        "# 6 mice for training, 2 mice for test and validation (one wound on each mice for test one for validation)\n",
        "features_train, labels_train, video_files_paths_train = create_dataset('train')\n",
        "features_test, labels_test, video_files_paths_test = create_dataset('test')\n",
        "features_valid, labels_valid, video_files_paths_valid = create_dataset('valid')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "dtJkK4qTAulC"
      },
      "outputs": [],
      "source": [
        "# labels to catogorical\n",
        "labels_train = keras.utils.to_categorical(labels_train)\n",
        "labels_test = keras.utils.to_categorical(labels_test)\n",
        "labels_valid = keras.utils.to_categorical(labels_valid)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "Bi-NDol3DHRV"
      },
      "outputs": [],
      "source": [
        "def create_convlstm_model():\n",
        "    # TimeDistributed is a wrapper to handle input of size five to maintain frames number for LSTM\n",
        "    # A Conv2D layer requires four dimensions: (batch_size, height, width, channels)\n",
        "    # TimeDistributed will require an additional dimension: (batch_size, frames, height, width, channels)\n",
        "\n",
        "    model = Sequential()\n",
        "\n",
        "    model.add(TimeDistributed(Conv2D(8, (3,3), activation='relu'), input_shape=(16, 128, 128, 3)))\n",
        "    model.add(TimeDistributed(MaxPooling2D((2,2), strides=(2,2))))\n",
        "    model.add(Dropout(0.3))\n",
        "\n",
        "    model.add(TimeDistributed(Conv2D(16, (3,3), activation='relu')))\n",
        "    model.add(TimeDistributed(MaxPooling2D((2,2), strides=(2,2))))\n",
        "    model.add(Dropout(0.3))\n",
        "\n",
        "    model.add(TimeDistributed(Conv2D(24, (3,3), activation='relu')))\n",
        "    model.add(TimeDistributed(MaxPooling2D((2,2), strides=(2,2))))\n",
        "    model.add(Dropout(0.3))\n",
        "\n",
        "    model.add(TimeDistributed(Conv2D(32, (3,3), activation='relu')))\n",
        "    model.add(TimeDistributed(MaxPooling2D((2,2), strides=(2,2))))\n",
        "    model.add(Dropout(0.3))\n",
        "\n",
        "    model.add(TimeDistributed(Flatten()))\n",
        "    model.add(Dropout(0.2))\n",
        "\n",
        "    model.add(LSTM(4, activation='relu', return_sequences=False))\n",
        "    model.add(Dropout(0.2))\n",
        "\n",
        "    model.add(Dense(32, activation='relu', kernel_initializer='he_uniform'))\n",
        "    model.add(Dropout(0.2))\n",
        "\n",
        "    model.add(Dense(16, activation='relu', kernel_initializer='he_uniform'))\n",
        "    model.add(Dropout(0.2))\n",
        "\n",
        "    # model.add(Dense(8, activation='relu', kernel_initializer='he_uniform'))\n",
        "\n",
        "    model.add(Dense(4, activation='relu', kernel_initializer='he_uniform'))\n",
        "\n",
        "    model.add(Dense(len(CLASSES_LIST), activation='softmax'))\n",
        "    model.summary(line_length = 100)\n",
        "    return model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "FbjNYI-0DY_v",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ec16dd3d-f0e4-48a1-9022-e83699738ee3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "____________________________________________________________________________________________________\n",
            " Layer (type)                                Output Shape                            Param #        \n",
            "====================================================================================================\n",
            " time_distributed (TimeDistributed)          (None, 16, 126, 126, 8)                 224            \n",
            "                                                                                                    \n",
            " time_distributed_1 (TimeDistributed)        (None, 16, 63, 63, 8)                   0              \n",
            "                                                                                                    \n",
            " dropout (Dropout)                           (None, 16, 63, 63, 8)                   0              \n",
            "                                                                                                    \n",
            " time_distributed_2 (TimeDistributed)        (None, 16, 61, 61, 16)                  1168           \n",
            "                                                                                                    \n",
            " time_distributed_3 (TimeDistributed)        (None, 16, 30, 30, 16)                  0              \n",
            "                                                                                                    \n",
            " dropout_1 (Dropout)                         (None, 16, 30, 30, 16)                  0              \n",
            "                                                                                                    \n",
            " time_distributed_4 (TimeDistributed)        (None, 16, 28, 28, 24)                  3480           \n",
            "                                                                                                    \n",
            " time_distributed_5 (TimeDistributed)        (None, 16, 14, 14, 24)                  0              \n",
            "                                                                                                    \n",
            " dropout_2 (Dropout)                         (None, 16, 14, 14, 24)                  0              \n",
            "                                                                                                    \n",
            " time_distributed_6 (TimeDistributed)        (None, 16, 12, 12, 32)                  6944           \n",
            "                                                                                                    \n",
            " time_distributed_7 (TimeDistributed)        (None, 16, 6, 6, 32)                    0              \n",
            "                                                                                                    \n",
            " dropout_3 (Dropout)                         (None, 16, 6, 6, 32)                    0              \n",
            "                                                                                                    \n",
            " time_distributed_8 (TimeDistributed)        (None, 16, 1152)                        0              \n",
            "                                                                                                    \n",
            " dropout_4 (Dropout)                         (None, 16, 1152)                        0              \n",
            "                                                                                                    \n",
            " lstm (LSTM)                                 (None, 4)                               18512          \n",
            "                                                                                                    \n",
            " dropout_5 (Dropout)                         (None, 4)                               0              \n",
            "                                                                                                    \n",
            " dense (Dense)                               (None, 32)                              160            \n",
            "                                                                                                    \n",
            " dropout_6 (Dropout)                         (None, 32)                              0              \n",
            "                                                                                                    \n",
            " dense_1 (Dense)                             (None, 16)                              528            \n",
            "                                                                                                    \n",
            " dropout_7 (Dropout)                         (None, 16)                              0              \n",
            "                                                                                                    \n",
            " dense_2 (Dense)                             (None, 4)                               68             \n",
            "                                                                                                    \n",
            " dense_3 (Dense)                             (None, 4)                               20             \n",
            "                                                                                                    \n",
            "====================================================================================================\n",
            "Total params: 31,104\n",
            "Trainable params: 31,104\n",
            "Non-trainable params: 0\n",
            "____________________________________________________________________________________________________\n",
            "Model Created Successfully!\n"
          ]
        }
      ],
      "source": [
        "# Construct the required convlstm model.\n",
        "convlstm_model = create_convlstm_model()\n",
        " \n",
        "# Display the success message. \n",
        "print(\"Model Created Successfully!\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "SCB3R7so-wHz"
      },
      "outputs": [],
      "source": [
        "# keras.utils.plot_model(convlstm_model, to_file = 'convlstm_model_structure_plot.png', show_shapes = True, show_layer_names = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "nMYwQOwyF3bd"
      },
      "outputs": [],
      "source": [
        "# keras.utils.plot_model(convlstm_model,\n",
        "#                          to_file = 'convlstm_model_structure_plot.png',\n",
        "#                          show_shapes = True,\n",
        "#                          show_layer_names = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vYC_6H0uGqW9",
        "outputId": "41f02d9b-3454-447c-de7d-9c6e50c468e2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/150\n",
            "30/30 [==============================] - 11s 198ms/step - loss: 1.3195 - accuracy: 0.3769 - val_loss: 1.3070 - val_accuracy: 0.3566\n",
            "Epoch 2/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 1.1947 - accuracy: 0.3775 - val_loss: 1.2612 - val_accuracy: 0.3566\n",
            "Epoch 3/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 1.1121 - accuracy: 0.3843 - val_loss: 1.1675 - val_accuracy: 0.4485\n",
            "Epoch 4/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 1.0334 - accuracy: 0.4152 - val_loss: 1.1307 - val_accuracy: 0.6176\n",
            "Epoch 5/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.9670 - accuracy: 0.4957 - val_loss: 1.1116 - val_accuracy: 0.6471\n",
            "Epoch 6/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.9283 - accuracy: 0.5353 - val_loss: 1.0320 - val_accuracy: 0.6213\n",
            "Epoch 7/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.8995 - accuracy: 0.5718 - val_loss: 1.0240 - val_accuracy: 0.5809\n",
            "Epoch 8/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.8653 - accuracy: 0.6021 - val_loss: 0.9619 - val_accuracy: 0.6618\n",
            "Epoch 9/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.8471 - accuracy: 0.6064 - val_loss: 1.0099 - val_accuracy: 0.6140\n",
            "Epoch 10/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.8198 - accuracy: 0.6368 - val_loss: 0.9393 - val_accuracy: 0.5956\n",
            "Epoch 11/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.8044 - accuracy: 0.6597 - val_loss: 0.8498 - val_accuracy: 0.7353\n",
            "Epoch 12/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.7802 - accuracy: 0.6627 - val_loss: 0.9825 - val_accuracy: 0.5625\n",
            "Epoch 13/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.7906 - accuracy: 0.6442 - val_loss: 0.9492 - val_accuracy: 0.6691\n",
            "Epoch 14/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.7564 - accuracy: 0.6807 - val_loss: 0.8782 - val_accuracy: 0.6581\n",
            "Epoch 15/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.7127 - accuracy: 0.6838 - val_loss: 0.8131 - val_accuracy: 0.6949\n",
            "Epoch 16/150\n",
            "30/30 [==============================] - 4s 143ms/step - loss: 0.7115 - accuracy: 0.7042 - val_loss: 0.7521 - val_accuracy: 0.7904\n",
            "Epoch 17/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.7095 - accuracy: 0.7054 - val_loss: 0.7846 - val_accuracy: 0.7426\n",
            "Epoch 18/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.7039 - accuracy: 0.6986 - val_loss: 0.7649 - val_accuracy: 0.7537\n",
            "Epoch 19/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.7180 - accuracy: 0.6949 - val_loss: 0.8232 - val_accuracy: 0.6360\n",
            "Epoch 20/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.6716 - accuracy: 0.7147 - val_loss: 0.7678 - val_accuracy: 0.7059\n",
            "Epoch 21/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.6485 - accuracy: 0.7191 - val_loss: 0.6743 - val_accuracy: 0.8015\n",
            "Epoch 22/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.6503 - accuracy: 0.7234 - val_loss: 0.7034 - val_accuracy: 0.7647\n",
            "Epoch 23/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.6338 - accuracy: 0.7321 - val_loss: 0.6831 - val_accuracy: 0.7831\n",
            "Epoch 24/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.6269 - accuracy: 0.7265 - val_loss: 0.7486 - val_accuracy: 0.6985\n",
            "Epoch 25/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.6128 - accuracy: 0.7389 - val_loss: 0.7050 - val_accuracy: 0.7353\n",
            "Epoch 26/150\n",
            "30/30 [==============================] - 4s 142ms/step - loss: 0.6009 - accuracy: 0.7308 - val_loss: 0.6851 - val_accuracy: 0.7390\n",
            "Epoch 27/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.6190 - accuracy: 0.7339 - val_loss: 0.6401 - val_accuracy: 0.7794\n",
            "Epoch 28/150\n",
            "30/30 [==============================] - 4s 142ms/step - loss: 0.5909 - accuracy: 0.7314 - val_loss: 0.6235 - val_accuracy: 0.7904\n",
            "Epoch 29/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.5755 - accuracy: 0.7420 - val_loss: 0.5876 - val_accuracy: 0.8125\n",
            "Epoch 30/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.5708 - accuracy: 0.7401 - val_loss: 0.5960 - val_accuracy: 0.8015\n",
            "Epoch 31/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.5658 - accuracy: 0.7376 - val_loss: 0.5862 - val_accuracy: 0.8015\n",
            "Epoch 32/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.5648 - accuracy: 0.7364 - val_loss: 0.5747 - val_accuracy: 0.8088\n",
            "Epoch 33/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5532 - accuracy: 0.7432 - val_loss: 0.5750 - val_accuracy: 0.8015\n",
            "Epoch 34/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5405 - accuracy: 0.7457 - val_loss: 0.5655 - val_accuracy: 0.7941\n",
            "Epoch 35/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5369 - accuracy: 0.7457 - val_loss: 0.5863 - val_accuracy: 0.7831\n",
            "Epoch 36/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.5403 - accuracy: 0.7413 - val_loss: 0.5734 - val_accuracy: 0.7978\n",
            "Epoch 37/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5335 - accuracy: 0.7500 - val_loss: 0.5040 - val_accuracy: 0.8346\n",
            "Epoch 38/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5227 - accuracy: 0.8119 - val_loss: 0.5329 - val_accuracy: 0.8346\n",
            "Epoch 39/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5280 - accuracy: 0.8014 - val_loss: 0.5303 - val_accuracy: 0.8309\n",
            "Epoch 40/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.5141 - accuracy: 0.8150 - val_loss: 0.5418 - val_accuracy: 0.8088\n",
            "Epoch 41/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.5131 - accuracy: 0.8175 - val_loss: 0.4961 - val_accuracy: 0.7868\n",
            "Epoch 42/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.5582 - accuracy: 0.7896 - val_loss: 0.6006 - val_accuracy: 0.7353\n",
            "Epoch 43/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.5934 - accuracy: 0.7525 - val_loss: 0.5916 - val_accuracy: 0.8015\n",
            "Epoch 44/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.5248 - accuracy: 0.8162 - val_loss: 0.5345 - val_accuracy: 0.8051\n",
            "Epoch 45/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.4976 - accuracy: 0.8205 - val_loss: 0.5222 - val_accuracy: 0.8235\n",
            "Epoch 46/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.4778 - accuracy: 0.8329 - val_loss: 0.4951 - val_accuracy: 0.8382\n",
            "Epoch 47/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.4896 - accuracy: 0.8292 - val_loss: 0.4916 - val_accuracy: 0.8382\n",
            "Epoch 48/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4893 - accuracy: 0.8187 - val_loss: 0.4859 - val_accuracy: 0.8309\n",
            "Epoch 49/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.4727 - accuracy: 0.8267 - val_loss: 0.5101 - val_accuracy: 0.8199\n",
            "Epoch 50/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.4775 - accuracy: 0.8249 - val_loss: 0.5013 - val_accuracy: 0.8235\n",
            "Epoch 51/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4600 - accuracy: 0.8478 - val_loss: 0.4746 - val_accuracy: 0.8419\n",
            "Epoch 52/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.4642 - accuracy: 0.8354 - val_loss: 0.4544 - val_accuracy: 0.8419\n",
            "Epoch 53/150\n",
            "30/30 [==============================] - 4s 142ms/step - loss: 0.4543 - accuracy: 0.8527 - val_loss: 0.4730 - val_accuracy: 0.8382\n",
            "Epoch 54/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4586 - accuracy: 0.8391 - val_loss: 0.4605 - val_accuracy: 0.8419\n",
            "Epoch 55/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4605 - accuracy: 0.8366 - val_loss: 0.4569 - val_accuracy: 0.8419\n",
            "Epoch 56/150\n",
            "30/30 [==============================] - 4s 144ms/step - loss: 0.4509 - accuracy: 0.8521 - val_loss: 0.4156 - val_accuracy: 0.8456\n",
            "Epoch 57/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4483 - accuracy: 0.8397 - val_loss: 0.4517 - val_accuracy: 0.8382\n",
            "Epoch 58/150\n",
            "30/30 [==============================] - 4s 143ms/step - loss: 0.4346 - accuracy: 0.8546 - val_loss: 0.4370 - val_accuracy: 0.8382\n",
            "Epoch 59/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.4315 - accuracy: 0.8595 - val_loss: 0.4546 - val_accuracy: 0.8419\n",
            "Epoch 60/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.4360 - accuracy: 0.8484 - val_loss: 0.4445 - val_accuracy: 0.8419\n",
            "Epoch 61/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4356 - accuracy: 0.8521 - val_loss: 0.4427 - val_accuracy: 0.8419\n",
            "Epoch 62/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.4273 - accuracy: 0.8540 - val_loss: 0.4561 - val_accuracy: 0.8346\n",
            "Epoch 63/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.4255 - accuracy: 0.8490 - val_loss: 0.4807 - val_accuracy: 0.8382\n",
            "Epoch 64/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.4222 - accuracy: 0.8564 - val_loss: 0.4357 - val_accuracy: 0.8199\n",
            "Epoch 65/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.4176 - accuracy: 0.8601 - val_loss: 0.4324 - val_accuracy: 0.8419\n",
            "Epoch 66/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3945 - accuracy: 0.8713 - val_loss: 0.4184 - val_accuracy: 0.8419\n",
            "Epoch 67/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3948 - accuracy: 0.8682 - val_loss: 0.4245 - val_accuracy: 0.8419\n",
            "Epoch 68/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.4170 - accuracy: 0.8657 - val_loss: 0.4374 - val_accuracy: 0.8419\n",
            "Epoch 69/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.4062 - accuracy: 0.8651 - val_loss: 0.4299 - val_accuracy: 0.8346\n",
            "Epoch 70/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3903 - accuracy: 0.8787 - val_loss: 0.4467 - val_accuracy: 0.8346\n",
            "Epoch 71/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3849 - accuracy: 0.8762 - val_loss: 0.4075 - val_accuracy: 0.8382\n",
            "Epoch 72/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.3864 - accuracy: 0.8694 - val_loss: 0.4110 - val_accuracy: 0.8382\n",
            "Epoch 73/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3885 - accuracy: 0.8713 - val_loss: 0.4207 - val_accuracy: 0.8346\n",
            "Epoch 74/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3755 - accuracy: 0.8800 - val_loss: 0.3988 - val_accuracy: 0.8419\n",
            "Epoch 75/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3900 - accuracy: 0.8639 - val_loss: 0.4159 - val_accuracy: 0.8162\n",
            "Epoch 76/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3790 - accuracy: 0.8756 - val_loss: 0.4012 - val_accuracy: 0.8346\n",
            "Epoch 77/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3903 - accuracy: 0.8657 - val_loss: 0.4246 - val_accuracy: 0.8346\n",
            "Epoch 78/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3674 - accuracy: 0.8787 - val_loss: 0.3981 - val_accuracy: 0.8382\n",
            "Epoch 79/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.3616 - accuracy: 0.8892 - val_loss: 0.3796 - val_accuracy: 0.8419\n",
            "Epoch 80/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.3553 - accuracy: 0.8818 - val_loss: 0.4039 - val_accuracy: 0.8456\n",
            "Epoch 81/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3470 - accuracy: 0.8892 - val_loss: 0.3950 - val_accuracy: 0.8456\n",
            "Epoch 82/150\n",
            "30/30 [==============================] - 4s 143ms/step - loss: 0.3305 - accuracy: 0.9016 - val_loss: 0.4183 - val_accuracy: 0.8419\n",
            "Epoch 83/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3335 - accuracy: 0.8942 - val_loss: 0.3895 - val_accuracy: 0.8235\n",
            "Epoch 84/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3266 - accuracy: 0.8960 - val_loss: 0.4221 - val_accuracy: 0.8382\n",
            "Epoch 85/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3215 - accuracy: 0.8998 - val_loss: 0.4030 - val_accuracy: 0.8456\n",
            "Epoch 86/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3254 - accuracy: 0.8973 - val_loss: 0.3702 - val_accuracy: 0.8419\n",
            "Epoch 87/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3096 - accuracy: 0.9022 - val_loss: 0.4422 - val_accuracy: 0.8162\n",
            "Epoch 88/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3364 - accuracy: 0.8936 - val_loss: 0.4176 - val_accuracy: 0.8088\n",
            "Epoch 89/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3630 - accuracy: 0.8756 - val_loss: 0.3695 - val_accuracy: 0.8456\n",
            "Epoch 90/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3309 - accuracy: 0.8899 - val_loss: 0.3865 - val_accuracy: 0.8382\n",
            "Epoch 91/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3991 - accuracy: 0.8620 - val_loss: 0.3989 - val_accuracy: 0.8346\n",
            "Epoch 92/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3341 - accuracy: 0.8911 - val_loss: 0.4419 - val_accuracy: 0.8456\n",
            "Epoch 93/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3191 - accuracy: 0.8923 - val_loss: 0.3869 - val_accuracy: 0.8125\n",
            "Epoch 94/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3076 - accuracy: 0.8998 - val_loss: 0.3936 - val_accuracy: 0.8382\n",
            "Epoch 95/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3236 - accuracy: 0.8979 - val_loss: 0.3751 - val_accuracy: 0.8382\n",
            "Epoch 96/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3161 - accuracy: 0.9004 - val_loss: 0.4089 - val_accuracy: 0.8419\n",
            "Epoch 97/150\n",
            "30/30 [==============================] - 4s 142ms/step - loss: 0.2949 - accuracy: 0.9097 - val_loss: 0.4189 - val_accuracy: 0.8419\n",
            "Epoch 98/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3013 - accuracy: 0.9066 - val_loss: 0.4491 - val_accuracy: 0.8419\n",
            "Epoch 99/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.3185 - accuracy: 0.8923 - val_loss: 0.4688 - val_accuracy: 0.7978\n",
            "Epoch 100/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.3139 - accuracy: 0.8998 - val_loss: 0.4086 - val_accuracy: 0.8199\n",
            "Epoch 101/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2959 - accuracy: 0.9028 - val_loss: 0.3931 - val_accuracy: 0.8162\n",
            "Epoch 102/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3039 - accuracy: 0.9059 - val_loss: 0.3711 - val_accuracy: 0.8419\n",
            "Epoch 103/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2894 - accuracy: 0.9084 - val_loss: 0.4328 - val_accuracy: 0.8309\n",
            "Epoch 104/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.3075 - accuracy: 0.9016 - val_loss: 0.3999 - val_accuracy: 0.8419\n",
            "Epoch 105/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2928 - accuracy: 0.9016 - val_loss: 0.3596 - val_accuracy: 0.8419\n",
            "Epoch 106/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2871 - accuracy: 0.9090 - val_loss: 0.4365 - val_accuracy: 0.8346\n",
            "Epoch 107/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2825 - accuracy: 0.9078 - val_loss: 0.4145 - val_accuracy: 0.8419\n",
            "Epoch 108/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2721 - accuracy: 0.9115 - val_loss: 0.4111 - val_accuracy: 0.8235\n",
            "Epoch 109/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2738 - accuracy: 0.9109 - val_loss: 0.4527 - val_accuracy: 0.8162\n",
            "Epoch 110/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2690 - accuracy: 0.9134 - val_loss: 0.4085 - val_accuracy: 0.8493\n",
            "Epoch 111/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2661 - accuracy: 0.9152 - val_loss: 0.4473 - val_accuracy: 0.8382\n",
            "Epoch 112/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2628 - accuracy: 0.9134 - val_loss: 0.3562 - val_accuracy: 0.8419\n",
            "Epoch 113/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.2688 - accuracy: 0.9146 - val_loss: 0.4094 - val_accuracy: 0.8162\n",
            "Epoch 114/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2744 - accuracy: 0.9090 - val_loss: 0.4111 - val_accuracy: 0.8199\n",
            "Epoch 115/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2898 - accuracy: 0.9041 - val_loss: 0.5300 - val_accuracy: 0.8125\n",
            "Epoch 116/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2776 - accuracy: 0.9078 - val_loss: 0.4392 - val_accuracy: 0.8162\n",
            "Epoch 117/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2999 - accuracy: 0.8960 - val_loss: 0.3901 - val_accuracy: 0.8162\n",
            "Epoch 118/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2772 - accuracy: 0.9140 - val_loss: 0.3527 - val_accuracy: 0.8419\n",
            "Epoch 119/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2454 - accuracy: 0.9245 - val_loss: 0.4001 - val_accuracy: 0.8199\n",
            "Epoch 120/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2755 - accuracy: 0.9066 - val_loss: 0.3949 - val_accuracy: 0.8235\n",
            "Epoch 121/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2651 - accuracy: 0.9134 - val_loss: 0.3938 - val_accuracy: 0.8419\n",
            "Epoch 122/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2565 - accuracy: 0.9158 - val_loss: 0.4376 - val_accuracy: 0.8125\n",
            "Epoch 123/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2582 - accuracy: 0.9177 - val_loss: 0.4371 - val_accuracy: 0.8235\n",
            "Epoch 124/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2979 - accuracy: 0.8973 - val_loss: 0.3577 - val_accuracy: 0.8419\n",
            "Epoch 125/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2768 - accuracy: 0.9090 - val_loss: 0.3488 - val_accuracy: 0.8382\n",
            "Epoch 126/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2705 - accuracy: 0.9146 - val_loss: 0.3201 - val_accuracy: 0.8676\n",
            "Epoch 127/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2549 - accuracy: 0.9140 - val_loss: 0.3679 - val_accuracy: 0.8456\n",
            "Epoch 128/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2447 - accuracy: 0.9239 - val_loss: 0.3468 - val_accuracy: 0.8199\n",
            "Epoch 129/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.2540 - accuracy: 0.9127 - val_loss: 0.3335 - val_accuracy: 0.8456\n",
            "Epoch 130/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2658 - accuracy: 0.9090 - val_loss: 0.4610 - val_accuracy: 0.7426\n",
            "Epoch 131/150\n",
            "30/30 [==============================] - 4s 136ms/step - loss: 0.2529 - accuracy: 0.9140 - val_loss: 0.3896 - val_accuracy: 0.8199\n",
            "Epoch 132/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2542 - accuracy: 0.9140 - val_loss: 0.3827 - val_accuracy: 0.8088\n",
            "Epoch 133/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.2525 - accuracy: 0.9165 - val_loss: 0.4386 - val_accuracy: 0.8162\n",
            "Epoch 134/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.2407 - accuracy: 0.9214 - val_loss: 0.3476 - val_accuracy: 0.8419\n",
            "Epoch 135/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2405 - accuracy: 0.9196 - val_loss: 0.5547 - val_accuracy: 0.8088\n",
            "Epoch 136/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2440 - accuracy: 0.9140 - val_loss: 0.4050 - val_accuracy: 0.8162\n",
            "Epoch 137/150\n",
            "30/30 [==============================] - 4s 137ms/step - loss: 0.2348 - accuracy: 0.9233 - val_loss: 0.4113 - val_accuracy: 0.8382\n",
            "Epoch 138/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2312 - accuracy: 0.9226 - val_loss: 0.4332 - val_accuracy: 0.8382\n",
            "Epoch 139/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2382 - accuracy: 0.9233 - val_loss: 0.3972 - val_accuracy: 0.8162\n",
            "Epoch 140/150\n",
            "30/30 [==============================] - 4s 141ms/step - loss: 0.2461 - accuracy: 0.9208 - val_loss: 0.4224 - val_accuracy: 0.8419\n",
            "Epoch 141/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2340 - accuracy: 0.9245 - val_loss: 0.3982 - val_accuracy: 0.8419\n",
            "Epoch 142/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2428 - accuracy: 0.9158 - val_loss: 0.4764 - val_accuracy: 0.8419\n",
            "Epoch 143/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2377 - accuracy: 0.9202 - val_loss: 0.3746 - val_accuracy: 0.8382\n",
            "Epoch 144/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2217 - accuracy: 0.9282 - val_loss: 0.3609 - val_accuracy: 0.8456\n",
            "Epoch 145/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.3112 - accuracy: 0.9016 - val_loss: 0.6232 - val_accuracy: 0.6985\n",
            "Epoch 146/150\n",
            "30/30 [==============================] - 4s 139ms/step - loss: 0.2983 - accuracy: 0.8936 - val_loss: 0.4448 - val_accuracy: 0.7721\n",
            "Epoch 147/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2458 - accuracy: 0.9158 - val_loss: 0.4513 - val_accuracy: 0.7978\n",
            "Epoch 148/150\n",
            "30/30 [==============================] - 4s 142ms/step - loss: 0.2298 - accuracy: 0.9214 - val_loss: 0.4658 - val_accuracy: 0.8015\n",
            "Epoch 149/150\n",
            "30/30 [==============================] - 4s 138ms/step - loss: 0.2354 - accuracy: 0.9202 - val_loss: 0.4200 - val_accuracy: 0.8382\n",
            "Epoch 150/150\n",
            "30/30 [==============================] - 4s 140ms/step - loss: 0.2284 - accuracy: 0.9239 - val_loss: 0.4401 - val_accuracy: 0.8125\n"
          ]
        }
      ],
      "source": [
        "# Create an Instance of Early Stopping Callback\n",
        "\n",
        "# https://stackoverflow.com/questions/53479007/how-to-setup-adaptive-learning-rate-in-keras\n",
        "# def adapt_learning_rate(epoch):\n",
        "#     print(0.0001 * epoch)\n",
        "#     return 0.0001 * epoch\n",
        "# my_lr_scheduler = keras.callbacks.LearningRateScheduler(adapt_learning_rate)\n",
        "\n",
        "early_stopping_callback = keras.callbacks.EarlyStopping(monitor = 'val_loss',\n",
        "                                                        patience = 10,\n",
        "                                                        mode = 'min',\n",
        "                                                        restore_best_weights = True)\n",
        "# Compile the model and specify loss function, optimizer and metrics values to the model\n",
        "convlstm_model.compile(loss = 'categorical_crossentropy',\n",
        "                       optimizer= keras.optimizers.Adam(0.0008, decay=1e-4),\n",
        "                       metrics = [\"accuracy\"])\n",
        "# Start training the model.\n",
        "convlstm_model_training_history = convlstm_model.fit(x = features_train,\n",
        "                                                     y = labels_train,\n",
        "                                                     epochs = 150,\n",
        "                                                     batch_size = 54,\n",
        "                                                     shuffle = True,\n",
        "                                                     validation_data = (features_valid, labels_valid)),\n",
        "                                                    #  callbacks = [early_stopping_callback])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LcpKhGqvIenv",
        "outputId": "9ffbe512-9a42-429a-f149-a8c30b121cca"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "9/9 [==============================] - 1s 41ms/step - loss: 1.1575 - accuracy: 0.6642\n",
            "\n",
            "\n",
            "Train accuracy: 92.636 % || Test accuracy: 66.421 % || Val accuracy: 81.250 %\n",
            "\n",
            "\n",
            "Train loss: 0.215 || Test loss: 1.157 || Val loss: 0.440\n"
          ]
        }
      ],
      "source": [
        "model_evaluation_history = convlstm_model.evaluate(features_test, labels_test)\n",
        "print('\\n')\n",
        "train_loss, train_acc = convlstm_model.evaluate(features_train, labels_train, verbose=0)\n",
        "test_loss, test_acc = convlstm_model.evaluate(features_test, labels_test, verbose=0)\n",
        "val_loss, val_acc = convlstm_model.evaluate(features_valid, labels_valid, verbose=0)\n",
        "\n",
        "print(f'Train accuracy: {train_acc*100:.3f} % || Test accuracy: {test_acc*100:.3f} % || Val accuracy: {val_acc*100:.3f} %')\n",
        "print('\\n')\n",
        "print(f'Train loss: {train_loss:.3f} || Test loss: {test_loss:.3f} || Val loss: {val_loss:.3f}')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "qsFw5KIzIum9"
      },
      "outputs": [],
      "source": [
        "# Get the loss and accuracy from model_evaluation_history.\n",
        "model_evaluation_loss, model_evaluation_accuracy = model_evaluation_history\n",
        " \n",
        "# Define the string date format.\n",
        "# Get the current Date and Time in a DateTime Object.\n",
        "# Convert the DateTime object to string according to the style mentioned in date_time_format string.\n",
        "date_time_format = '%Y_%m_%d__%H_%M_%S'\n",
        "current_date_time_dt = dt.datetime.now()\n",
        "current_date_time_string = dt.datetime.strftime(current_date_time_dt, date_time_format)\n",
        " \n",
        "# Define a useful name for our model to make it easy for us while navigating through multiple saved models.\n",
        "model_file_name = f'convlstm_model___Date_Time_{current_date_time_string}___Loss_{model_evaluation_loss}___Accuracy_{model_evaluation_accuracy}.h5'\n",
        " \n",
        "# Change dir\n",
        "gdrive_path = '/content/gdrive' + '/My Drive/247/Saved_models/'\n",
        "os.chdir(gdrive_path)\n",
        "# Create a floder for the model files\n",
        "!mkdir -p convlstm_{current_date_time_string}\n",
        "# Save your Model.\n",
        "convlstm_model.save('convlstm_' + str(current_date_time_string) + '/' + model_file_name)\n",
        "# Save model weights\n",
        "convlstm_model.save_weights('convlstm_' + str(current_date_time_string) + '/' + 'weights')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "Z0eAobhSIz_P"
      },
      "outputs": [],
      "source": [
        "def plot_metric(model_training_history, metric_name_1, metric_name_2, plot_name):\n",
        "    '''\n",
        "    This function will plot the metrics passed to it in a graph.\n",
        "    Args:\n",
        "        model_training_history: A history object containing a record of training and validation \n",
        "                                loss values and metrics values at successive epochs\n",
        "        metric_name_1:          The name of the first metric that needs to be plotted in the graph.\n",
        "        metric_name_2:          The name of the second metric that needs to be plotted in the graph.\n",
        "        plot_name:              The title of the graph.\n",
        "    '''\n",
        "    \n",
        "    # Get metric values using metric names as identifiers.\n",
        "    metric_value_1 = model_training_history[0].history[metric_name_1]\n",
        "    metric_value_2 = model_training_history[0].history[metric_name_2]\n",
        "    \n",
        "    # Construct a range object which will be used as x-axis (horizontal plane) of the graph.\n",
        "    epochs = range(len(metric_value_1))\n",
        "\n",
        "    # Plot the Graph.\n",
        "    plt.plot(epochs, metric_value_1, 'blue', label = metric_name_1)\n",
        "    plt.plot(epochs, metric_value_2, 'red', label = metric_name_2)\n",
        "\n",
        "    # Add title to the plot.\n",
        "    plt.title(str(plot_name))\n",
        "\n",
        "    # Add legend to the plot.\n",
        "    plt.legend()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "18RtBjI9I5lE",
        "outputId": "e2db18a3-172a-4895-e92a-4a4b9536e6ba"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plot_metric(convlstm_model_training_history, 'loss', 'val_loss', 'Total Loss vs Total Validation Loss')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "02amk7TXI_OS",
        "outputId": "ccbdfb79-aa23-42b5-e00b-b5f1172fc693"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plot_metric(convlstm_model_training_history, 'accuracy', 'val_accuracy', 'Total Accuracy vs Total Validation Accuracy') "
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "machine_shape": "hm",
      "name": "2D_CNN+LSTM.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyP73nhR19Qdx0+f9sPTkg7Y",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}