chapter_8.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TensorFlow version: 2.18.0\n",
      "GPU Available:  [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]\n",
      "GPU /physical_device:GPU:0 memory limit: 12.00 GB\n",
      "Default GPU: /device:GPU:0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "I0000 00:00:1736413355.271950  184494 gpu_device.cc:2022] Created device /device:GPU:0 with 9571 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3080, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import keras_tuner as kt\n",
    "import numpy as np\n",
    "\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "from matplotlib import pyplot as plt\n",
    "from tensorflow.keras.models import load_model\n",
    "\n",
    "# GPU 사용 가능 여부 확인\n",
    "print(\"TensorFlow version:\", tf.__version__)\n",
    "print(\"GPU Available: \", tf.config.list_physical_devices('GPU'))\n",
    "\n",
    "# GPU 메모리 증가 설정\n",
    "gpus = tf.config.list_physical_devices('GPU')\n",
    "if gpus:\n",
    "    try:\n",
    "        for gpu in gpus:\n",
    "            tf.config.experimental.set_memory_growth(gpu, True)\n",
    "            # Get GPU memory info using nvidia-smi\n",
    "            import subprocess\n",
    "            def get_gpu_memory():\n",
    "                command = \"nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits\"\n",
    "                memory_total = subprocess.check_output(command.split()).decode('ascii').strip().split('\\n')[0]\n",
    "                return int(memory_total)\n",
    "            \n",
    "            total_memory = get_gpu_memory()\n",
    "            print(f'GPU {gpu.name} memory limit: {total_memory/1024:.2f} GB')\n",
    "    except (RuntimeError, FileNotFoundError, subprocess.CalledProcessError) as e:\n",
    "        print(f\"Error getting GPU info: {e}\")\n",
    "\n",
    "# GPU 장치 설정\n",
    "try:\n",
    "    print('Default GPU: {}'.format(tf.test.gpu_device_name()))\n",
    "except Exception as e:\n",
    "    print(f\"Error while finding GPUs: {e}\")\n",
    "    print(f\"Available GPUs: {gpus[0].name}\")\n",
    "\n",
    "keras.mixed_precision.set_global_policy('mixed_float16')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Not Using a Pre-trained Model\n",
    "### Training ConvNet from Scratch\n",
    "#### Preparing dogs-vs-cats dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil, pathlib\n",
    "\n",
    "# directory path of the original dataset\n",
    "original_dir = pathlib.Path('./dogs-vs-cats/train')\n",
    "new_base_dir = pathlib.Path('./dogs-vs-cats_small')\n",
    "\n",
    "# organize the dataset into train, validation, and test sets\n",
    "def make_subset(subset_name, start_index, end_index):\n",
    "    for category in (\"dog\", \"cat\"):\n",
    "        dir = new_base_dir / subset_name / category\n",
    "        pathlib.Path(dir).mkdir(parents=True, exist_ok=True)\n",
    "        \n",
    "        fnames = [f\"{category}.{i}.jpg\" for i in range(start_index, end_index)]\n",
    "        \n",
    "        for fname in fnames:\n",
    "            shutil.copyfile(src = original_dir / fname,\n",
    "                            dst = dir / fname)\n",
    "\n",
    "make_subset(\"train\", 0, 1000)\n",
    "make_subset(\"validation\", 1000, 1500)\n",
    "make_subset(\"test\", 1500, 2500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2000 files belonging to 2 classes.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "I0000 00:00:1736413377.625535  184494 gpu_device.cc:2022] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 9571 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3080, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1000 files belonging to 2 classes.\n",
      "Found 2000 files belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.utils import image_dataset_from_directory\n",
    "\n",
    "# loading the dataset into the memory in batches\n",
    "train_dataset = image_dataset_from_directory(\n",
    "    directory = new_base_dir / 'train',\n",
    "    image_size = (180, 180),\n",
    "    batch_size = 32\n",
    ")\n",
    "\n",
    "validation_dataset = image_dataset_from_directory(\n",
    "    directory = new_base_dir / 'validation',\n",
    "    image_size = (180, 180),\n",
    "    batch_size = 32\n",
    ")\n",
    "\n",
    "test_dataset = image_dataset_from_directory(\n",
    "    directory = new_base_dir / 'test',\n",
    "    image_size = (180, 180),\n",
    "    batch_size = 32\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data batch shape: (32, 180, 180, 3)\n",
      "labels batch shape: (32,)\n"
     ]
    }
   ],
   "source": [
    "# checking the shape of the data and labels\n",
    "for data_batch, labels_batch in train_dataset:\n",
    "    print('data batch shape:', data_batch.shape)\n",
    "    print('labels batch shape:', labels_batch.shape)\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preparing the Scratch-Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"convnet_from_scratch\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"convnet_from_scratch\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)   │           <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12544</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │        <span style=\"color: #00af00; text-decoration-color: #00af00\">12,545</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast (\u001b[38;5;33mCast\u001b[0m)                     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling (\u001b[38;5;33mRescaling\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m32\u001b[0m)   │           \u001b[38;5;34m896\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m18,496\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │        \u001b[38;5;34m73,856\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m295,168\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12544\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │        \u001b[38;5;34m12,545\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,041</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m991,041\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,041</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m991,041\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# designing and compiling the model\n",
    "convnet_inputs = keras.Input(shape=(180, 180, 3))\n",
    "\n",
    "convnet_layer = layers.Rescaling(1.0/255)(convnet_inputs)\n",
    "convnet_layer = layers.Conv2D(filters = 32, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 64, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 128, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 256, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 256, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.Flatten()(convnet_layer)\n",
    "\n",
    "convnet_outputs = layers.Dense(1, activation='sigmoid')(convnet_layer)\n",
    "convnet_model = keras.Model(inputs = convnet_inputs, outputs = convnet_outputs, name=\"convnet_from_scratch\")\n",
    "\n",
    "convnet_model.summary()\n",
    "\n",
    "convnet_model.compile(optimizer='rmsprop',\n",
    "            loss='binary_crossentropy',\n",
    "            metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# defining a callback to save the best model\n",
    "def checkpoint_save(filepath):\n",
    "    checkpoint = keras.callbacks.ModelCheckpoint(\n",
    "        filepath = filepath,\n",
    "        save_best_only = True,\n",
    "        monitor = 'val_loss',\n",
    "    )\n",
    "    return checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "I0000 00:00:1736328661.977542  921240 service.cc:148] XLA service 0x7f5de000b460 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
      "I0000 00:00:1736328661.977615  921240 service.cc:156]   StreamExecutor device (0): NVIDIA GeForce RTX 3080, Compute Capability 8.6\n",
      "2025-01-08 18:31:02.017609: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:268] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
      "I0000 00:00:1736328662.190641  921240 cuda_dnn.cc:529] Loaded cuDNN version 90600\n",
      "2025-01-08 18:31:04.122478: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng11{k2=3,k3=0} for conv (f16[32,178,178,32]{3,2,1,0}, u8[0]{0}) custom-call(f16[32,180,180,3]{3,2,1,0}, f16[32,3,3,3]{3,2,1,0}, f16[32]{0}), window={size=3x3}, dim_labels=b01f_o01i->b01f, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"cudnn_conv_backend_config\":{\"activation_mode\":\"kNone\",\"conv_result_scale\":1,\"leakyrelu_alpha\":0,\"side_input_scale\":0},\"force_earliest_schedule\":false,\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[]} is taking a while...\n",
      "2025-01-08 18:31:04.652492: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.531078578s\n",
      "Trying algorithm eng11{k2=3,k3=0} for conv (f16[32,178,178,32]{3,2,1,0}, u8[0]{0}) custom-call(f16[32,180,180,3]{3,2,1,0}, f16[32,3,3,3]{3,2,1,0}, f16[32]{0}), window={size=3x3}, dim_labels=b01f_o01i->b01f, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"cudnn_conv_backend_config\":{\"activation_mode\":\"kNone\",\"conv_result_scale\":1,\"leakyrelu_alpha\":0,\"side_input_scale\":0},\"force_earliest_schedule\":false,\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[]} is taking a while...\n",
      "2025-01-08 18:31:08.736533: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng19{k2=3} for conv (f16[128,3,3,64]{3,2,1,0}, u8[0]{0}) custom-call(f16[32,43,43,64]{3,2,1,0}, f16[32,41,41,128]{3,2,1,0}), window={size=3x3}, dim_labels=b01f_o01i->b01f, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"cudnn_conv_backend_config\":{\"activation_mode\":\"kNone\",\"conv_result_scale\":1,\"leakyrelu_alpha\":0,\"side_input_scale\":0},\"force_earliest_schedule\":false,\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[]} is taking a while...\n",
      "2025-01-08 18:31:10.025494: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 2.289516606s\n",
      "Trying algorithm eng19{k2=3} for conv (f16[128,3,3,64]{3,2,1,0}, u8[0]{0}) custom-call(f16[32,43,43,64]{3,2,1,0}, f16[32,41,41,128]{3,2,1,0}), window={size=3x3}, dim_labels=b01f_o01i->b01f, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"cudnn_conv_backend_config\":{\"activation_mode\":\"kNone\",\"conv_result_scale\":1,\"leakyrelu_alpha\":0,\"side_input_scale\":0},\"force_earliest_schedule\":false,\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[]} is taking a while...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m15/63\u001b[0m \u001b[32m━━━━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 12ms/step - accuracy: 0.4685 - loss: 0.7158"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "I0000 00:00:1736328671.606423  921240 device_compiler.h:188] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 87ms/step - accuracy: 0.4886 - loss: 0.7028 - val_accuracy: 0.5000 - val_loss: 0.6983\n",
      "Epoch 2/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.4618 - loss: 0.7056 - val_accuracy: 0.5000 - val_loss: 0.6927\n",
      "Epoch 3/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.5346 - loss: 0.7091 - val_accuracy: 0.5340 - val_loss: 0.6784\n",
      "Epoch 4/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - accuracy: 0.5796 - loss: 0.6849 - val_accuracy: 0.6010 - val_loss: 0.6491\n",
      "Epoch 5/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.6284 - loss: 0.6714 - val_accuracy: 0.6170 - val_loss: 0.6637\n",
      "Epoch 6/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.6465 - loss: 0.6218 - val_accuracy: 0.6320 - val_loss: 0.6295\n",
      "Epoch 7/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.6875 - loss: 0.6032 - val_accuracy: 0.6660 - val_loss: 0.6285\n",
      "Epoch 8/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - accuracy: 0.7155 - loss: 0.5645 - val_accuracy: 0.6800 - val_loss: 0.5773\n",
      "Epoch 9/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.7269 - loss: 0.5467 - val_accuracy: 0.7000 - val_loss: 0.5682\n",
      "Epoch 10/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - accuracy: 0.7530 - loss: 0.5317 - val_accuracy: 0.7290 - val_loss: 0.5578\n",
      "Epoch 11/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7816 - loss: 0.4824 - val_accuracy: 0.7150 - val_loss: 0.5529\n",
      "Epoch 12/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.8109 - loss: 0.4193 - val_accuracy: 0.7000 - val_loss: 0.6641\n",
      "Epoch 13/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.8362 - loss: 0.3821 - val_accuracy: 0.7290 - val_loss: 0.6395\n",
      "Epoch 14/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.8624 - loss: 0.3330 - val_accuracy: 0.7260 - val_loss: 0.7560\n",
      "Epoch 15/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.8643 - loss: 0.3018 - val_accuracy: 0.7110 - val_loss: 0.7374\n",
      "Epoch 16/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.8923 - loss: 0.2663 - val_accuracy: 0.7400 - val_loss: 0.7282\n",
      "Epoch 17/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - accuracy: 0.9376 - loss: 0.1691 - val_accuracy: 0.7140 - val_loss: 1.2693\n",
      "Epoch 18/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9365 - loss: 0.1663 - val_accuracy: 0.7300 - val_loss: 0.9898\n",
      "Epoch 19/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9653 - loss: 0.1094 - val_accuracy: 0.7370 - val_loss: 0.9951\n",
      "Epoch 20/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.9632 - loss: 0.1052 - val_accuracy: 0.7070 - val_loss: 1.2856\n",
      "Epoch 21/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9616 - loss: 0.1045 - val_accuracy: 0.7380 - val_loss: 1.2246\n",
      "Epoch 22/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.9703 - loss: 0.0752 - val_accuracy: 0.7410 - val_loss: 1.2332\n",
      "Epoch 23/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.9910 - loss: 0.0412 - val_accuracy: 0.6890 - val_loss: 2.0834\n",
      "Epoch 24/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - accuracy: 0.9740 - loss: 0.0943 - val_accuracy: 0.7440 - val_loss: 1.5402\n",
      "Epoch 25/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.9847 - loss: 0.0572 - val_accuracy: 0.7040 - val_loss: 1.5618\n",
      "Epoch 26/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9913 - loss: 0.0339 - val_accuracy: 0.6720 - val_loss: 2.5006\n",
      "Epoch 27/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9598 - loss: 0.1658 - val_accuracy: 0.7460 - val_loss: 1.9391\n",
      "Epoch 28/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - accuracy: 0.9799 - loss: 0.0598 - val_accuracy: 0.7350 - val_loss: 2.0876\n",
      "Epoch 29/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9885 - loss: 0.0325 - val_accuracy: 0.7310 - val_loss: 2.2533\n",
      "Epoch 30/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.9663 - loss: 0.1043 - val_accuracy: 0.7180 - val_loss: 2.2433\n"
     ]
    }
   ],
   "source": [
    "# training the model\n",
    "conv_history = convnet_model.fit(\n",
    "    train_dataset,\n",
    "    epochs = 30,\n",
    "    validation_data = validation_dataset,\n",
    "    callbacks = checkpoint_save(\"convnet_from_scratch.keras\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualizing Model-Training History\n",
    "##### Defining plotting function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# defining a function to plot the training history\n",
    "def plot_history(history):\n",
    "    accuracy = history.history['accuracy']\n",
    "    val_accuracy = history.history['val_accuracy']\n",
    "    loss = history.history['loss']\n",
    "    val_loss = history.history['val_loss']\n",
    "\n",
    "    fig = plt.figure(figsize=(10, 5))\n",
    "    ax1 = fig.add_subplot(1, 2, 1)\n",
    "    ax2 = fig.add_subplot(1, 2, 2)\n",
    "\n",
    "    ax1.plot(accuracy, label='accuracy')\n",
    "    ax1.plot(val_accuracy, label='val_accuracy')\n",
    "    ax1.legend()\n",
    "\n",
    "    ax2.plot(loss, label='loss')\n",
    "    ax2.plot(val_loss, label='val_loss')\n",
    "    ax2.legend()\n",
    "\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Plotting and visualizing the training history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the training history\n",
    "plot_history(conv_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 17ms/step - accuracy: 0.7168 - loss: 0.5961\n",
      "Test accuracy: 0.707\n"
     ]
    }
   ],
   "source": [
    "test_model = keras.models.load_model(\"convnet_from_scratch.keras\")\n",
    "test_loss, test_acc = test_model.evaluate(test_dataset)\n",
    "print(f\"Test accuracy: {test_acc:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# designing the data augmentation pipeline\n",
    "data_augmentation = keras.Sequential([\n",
    "    layers.RandomFlip(\"horizontal\"),\n",
    "    layers.RandomRotation(0.1),\n",
    "    layers.RandomZoom(0.2)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# visualizing the augmented images\n",
    "plt.figure(figsize=(10, 10))\n",
    "\n",
    "for images, _ in train_dataset.take(1):\n",
    "    for i in range(9):\n",
    "        ax = plt.subplot(3, 3, i + 1)\n",
    "        augmented_image = data_augmentation(images)\n",
    "        plt.imshow(augmented_image[0].numpy().astype(\"uint8\"))\n",
    "        plt.axis('off')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"convnet_from_scratch_augmented\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"convnet_from_scratch_augmented\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ sequential (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)   │           <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12544</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12544</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │        <span style=\"color: #00af00; text-decoration-color: #00af00\">12,545</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_2 (\u001b[38;5;33mCast\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ sequential (\u001b[38;5;33mSequential\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling_1 (\u001b[38;5;33mRescaling\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m32\u001b[0m)   │           \u001b[38;5;34m896\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_6 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m18,496\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_7 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │        \u001b[38;5;34m73,856\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_6 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_8 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m295,168\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_7 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_9 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12544\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12544\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │        \u001b[38;5;34m12,545\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,041</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m991,041\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,041</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m991,041\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# designing and compiling the model with data augmentation\n",
    "convnet_inputs = keras.Input(shape=(180, 180, 3))\n",
    "\n",
    "convnet_layer = data_augmentation(convnet_inputs)\n",
    "convnet_layer = layers.Rescaling(1.0/255)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 32, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 64, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 128, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 256, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.MaxPooling2D(pool_size = 2)(convnet_layer)\n",
    "convnet_layer = layers.Conv2D(filters = 256, kernel_size = 3, activation='relu')(convnet_layer)\n",
    "convnet_layer = layers.Flatten()(convnet_layer)\n",
    "convnet_layer = layers.Dropout(0.5)(convnet_layer)\n",
    "convnet_outputs = layers.Dense(1, activation='sigmoid')(convnet_layer)\n",
    "convnet_model = keras.Model(inputs = convnet_inputs, outputs = convnet_outputs, name=\"convnet_from_scratch_augmented\")\n",
    "\n",
    "convnet_model.summary()\n",
    "\n",
    "convnet_model.compile(optimizer='rmsprop',\n",
    "            loss='binary_crossentropy',\n",
    "            metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 48ms/step - accuracy: 0.5098 - loss: 0.7219 - val_accuracy: 0.5000 - val_loss: 0.6950\n",
      "Epoch 2/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - accuracy: 0.4994 - loss: 0.6964 - val_accuracy: 0.5350 - val_loss: 0.6844\n",
      "Epoch 3/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 33ms/step - accuracy: 0.5154 - loss: 0.7123 - val_accuracy: 0.5240 - val_loss: 0.6757\n",
      "Epoch 4/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.5760 - loss: 0.6732 - val_accuracy: 0.5530 - val_loss: 0.9081\n",
      "Epoch 5/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 30ms/step - accuracy: 0.6233 - loss: 0.6620 - val_accuracy: 0.6260 - val_loss: 0.6298\n",
      "Epoch 6/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.6534 - loss: 0.6350 - val_accuracy: 0.6320 - val_loss: 0.6347\n",
      "Epoch 7/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.6555 - loss: 0.6258 - val_accuracy: 0.6880 - val_loss: 0.5821\n",
      "Epoch 8/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.6612 - loss: 0.6180 - val_accuracy: 0.6830 - val_loss: 0.5828\n",
      "Epoch 9/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.6759 - loss: 0.6049 - val_accuracy: 0.6610 - val_loss: 0.5893\n",
      "Epoch 10/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.6859 - loss: 0.5816 - val_accuracy: 0.7010 - val_loss: 0.5779\n",
      "Epoch 11/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.7015 - loss: 0.5843 - val_accuracy: 0.6690 - val_loss: 0.5839\n",
      "Epoch 12/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.7198 - loss: 0.5788 - val_accuracy: 0.6470 - val_loss: 0.6149\n",
      "Epoch 13/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.7035 - loss: 0.5777 - val_accuracy: 0.5660 - val_loss: 1.0974\n",
      "Epoch 14/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.7068 - loss: 0.6236 - val_accuracy: 0.6290 - val_loss: 0.7082\n",
      "Epoch 15/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 33ms/step - accuracy: 0.7297 - loss: 0.5270 - val_accuracy: 0.6130 - val_loss: 0.7959\n",
      "Epoch 16/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.7301 - loss: 0.5552 - val_accuracy: 0.7160 - val_loss: 0.5839\n",
      "Epoch 17/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.7195 - loss: 0.5388 - val_accuracy: 0.7090 - val_loss: 0.5699\n",
      "Epoch 18/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.7440 - loss: 0.5205 - val_accuracy: 0.7370 - val_loss: 0.5137\n",
      "Epoch 19/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - accuracy: 0.7590 - loss: 0.5241 - val_accuracy: 0.7490 - val_loss: 0.5021\n",
      "Epoch 20/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.7646 - loss: 0.4853 - val_accuracy: 0.7420 - val_loss: 0.5346\n",
      "Epoch 21/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.7695 - loss: 0.4924 - val_accuracy: 0.7690 - val_loss: 0.4945\n",
      "Epoch 22/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.7646 - loss: 0.4552 - val_accuracy: 0.7230 - val_loss: 0.5564\n",
      "Epoch 23/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.7732 - loss: 0.4802 - val_accuracy: 0.7370 - val_loss: 0.5635\n",
      "Epoch 24/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.7845 - loss: 0.4669 - val_accuracy: 0.7530 - val_loss: 0.5145\n",
      "Epoch 25/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.7844 - loss: 0.4593 - val_accuracy: 0.7770 - val_loss: 0.4786\n",
      "Epoch 26/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.7894 - loss: 0.4457 - val_accuracy: 0.7630 - val_loss: 0.5177\n",
      "Epoch 27/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.7827 - loss: 0.4740 - val_accuracy: 0.7870 - val_loss: 0.4545\n",
      "Epoch 28/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.8004 - loss: 0.4245 - val_accuracy: 0.7950 - val_loss: 0.5085\n",
      "Epoch 29/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8061 - loss: 0.4056 - val_accuracy: 0.7720 - val_loss: 0.4856\n",
      "Epoch 30/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8177 - loss: 0.4199 - val_accuracy: 0.7810 - val_loss: 0.4911\n",
      "Epoch 31/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.8077 - loss: 0.4332 - val_accuracy: 0.6960 - val_loss: 0.7580\n",
      "Epoch 32/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.7943 - loss: 0.4200 - val_accuracy: 0.7970 - val_loss: 0.4576\n",
      "Epoch 33/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.8230 - loss: 0.4002 - val_accuracy: 0.6930 - val_loss: 0.6563\n",
      "Epoch 34/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.8106 - loss: 0.4280 - val_accuracy: 0.7720 - val_loss: 0.5209\n",
      "Epoch 35/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.8379 - loss: 0.3946 - val_accuracy: 0.7210 - val_loss: 0.5755\n",
      "Epoch 36/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - accuracy: 0.8151 - loss: 0.3873 - val_accuracy: 0.7730 - val_loss: 0.5347\n",
      "Epoch 37/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.8345 - loss: 0.3777 - val_accuracy: 0.7970 - val_loss: 0.6091\n",
      "Epoch 38/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.8538 - loss: 0.3461 - val_accuracy: 0.8150 - val_loss: 0.4981\n",
      "Epoch 39/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - accuracy: 0.8517 - loss: 0.3517 - val_accuracy: 0.7500 - val_loss: 0.8175\n",
      "Epoch 40/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.8532 - loss: 0.3647 - val_accuracy: 0.7940 - val_loss: 0.4965\n",
      "Epoch 41/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8581 - loss: 0.3338 - val_accuracy: 0.7820 - val_loss: 0.6706\n",
      "Epoch 42/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.8608 - loss: 0.3307 - val_accuracy: 0.7550 - val_loss: 0.7332\n",
      "Epoch 43/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step - accuracy: 0.8635 - loss: 0.3495 - val_accuracy: 0.7820 - val_loss: 0.6455\n",
      "Epoch 44/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.8533 - loss: 0.3224 - val_accuracy: 0.7740 - val_loss: 0.7993\n",
      "Epoch 45/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - accuracy: 0.8577 - loss: 0.3609 - val_accuracy: 0.7950 - val_loss: 0.7111\n",
      "Epoch 46/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8552 - loss: 0.3369 - val_accuracy: 0.7950 - val_loss: 0.6465\n",
      "Epoch 47/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8486 - loss: 0.3540 - val_accuracy: 0.7460 - val_loss: 0.7445\n",
      "Epoch 48/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.8583 - loss: 0.3313 - val_accuracy: 0.7780 - val_loss: 0.7000\n",
      "Epoch 49/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8796 - loss: 0.2903 - val_accuracy: 0.8270 - val_loss: 0.4505\n",
      "Epoch 50/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8831 - loss: 0.2821 - val_accuracy: 0.7710 - val_loss: 0.8782\n",
      "Epoch 51/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - accuracy: 0.8791 - loss: 0.2906 - val_accuracy: 0.8090 - val_loss: 0.4818\n",
      "Epoch 52/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.8922 - loss: 0.2864 - val_accuracy: 0.7940 - val_loss: 0.5660\n",
      "Epoch 53/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.8793 - loss: 0.2866 - val_accuracy: 0.8180 - val_loss: 0.6593\n",
      "Epoch 54/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.8822 - loss: 0.3120 - val_accuracy: 0.7820 - val_loss: 0.7994\n",
      "Epoch 55/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.8819 - loss: 0.2903 - val_accuracy: 0.8010 - val_loss: 0.6023\n",
      "Epoch 56/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9002 - loss: 0.2685 - val_accuracy: 0.8330 - val_loss: 0.4384\n",
      "Epoch 57/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9063 - loss: 0.2496 - val_accuracy: 0.7980 - val_loss: 0.6569\n",
      "Epoch 58/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.8908 - loss: 0.2637 - val_accuracy: 0.8140 - val_loss: 0.5578\n",
      "Epoch 59/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9127 - loss: 0.2419 - val_accuracy: 0.8200 - val_loss: 0.5247\n",
      "Epoch 60/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9039 - loss: 0.2500 - val_accuracy: 0.8150 - val_loss: 0.6920\n",
      "Epoch 61/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - accuracy: 0.9027 - loss: 0.2562 - val_accuracy: 0.8170 - val_loss: 0.7455\n",
      "Epoch 62/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9129 - loss: 0.2413 - val_accuracy: 0.8270 - val_loss: 0.6430\n",
      "Epoch 63/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9055 - loss: 0.2525 - val_accuracy: 0.8010 - val_loss: 0.7033\n",
      "Epoch 64/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9035 - loss: 0.2473 - val_accuracy: 0.8370 - val_loss: 0.5373\n",
      "Epoch 65/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9157 - loss: 0.2105 - val_accuracy: 0.8170 - val_loss: 0.5285\n",
      "Epoch 66/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9150 - loss: 0.2085 - val_accuracy: 0.8170 - val_loss: 0.6823\n",
      "Epoch 67/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9002 - loss: 0.2469 - val_accuracy: 0.7320 - val_loss: 1.5433\n",
      "Epoch 68/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9163 - loss: 0.2265 - val_accuracy: 0.8110 - val_loss: 0.7141\n",
      "Epoch 69/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9115 - loss: 0.2163 - val_accuracy: 0.8110 - val_loss: 0.6743\n",
      "Epoch 70/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9009 - loss: 0.2364 - val_accuracy: 0.7590 - val_loss: 0.8396\n",
      "Epoch 71/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9231 - loss: 0.2003 - val_accuracy: 0.8150 - val_loss: 0.6585\n",
      "Epoch 72/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9015 - loss: 0.2729 - val_accuracy: 0.8170 - val_loss: 0.6040\n",
      "Epoch 73/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9258 - loss: 0.2002 - val_accuracy: 0.8500 - val_loss: 0.4888\n",
      "Epoch 74/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9362 - loss: 0.1721 - val_accuracy: 0.8400 - val_loss: 0.5786\n",
      "Epoch 75/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9196 - loss: 0.2273 - val_accuracy: 0.7950 - val_loss: 0.6150\n",
      "Epoch 76/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9181 - loss: 0.2076 - val_accuracy: 0.8480 - val_loss: 0.5639\n",
      "Epoch 77/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9394 - loss: 0.1752 - val_accuracy: 0.8240 - val_loss: 0.6957\n",
      "Epoch 78/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9157 - loss: 0.2397 - val_accuracy: 0.8320 - val_loss: 0.7989\n",
      "Epoch 79/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9398 - loss: 0.1832 - val_accuracy: 0.6820 - val_loss: 1.5172\n",
      "Epoch 80/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9069 - loss: 0.2648 - val_accuracy: 0.8420 - val_loss: 0.6358\n",
      "Epoch 81/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - accuracy: 0.9309 - loss: 0.1915 - val_accuracy: 0.7680 - val_loss: 0.9869\n",
      "Epoch 82/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9233 - loss: 0.2275 - val_accuracy: 0.8200 - val_loss: 0.7039\n",
      "Epoch 83/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9417 - loss: 0.1522 - val_accuracy: 0.8410 - val_loss: 0.6058\n",
      "Epoch 84/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 26ms/step - accuracy: 0.9350 - loss: 0.1679 - val_accuracy: 0.8250 - val_loss: 0.7986\n",
      "Epoch 85/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - accuracy: 0.9239 - loss: 0.2118 - val_accuracy: 0.8500 - val_loss: 0.5902\n",
      "Epoch 86/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - accuracy: 0.9212 - loss: 0.1810 - val_accuracy: 0.8310 - val_loss: 0.7687\n",
      "Epoch 87/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step - accuracy: 0.9324 - loss: 0.1935 - val_accuracy: 0.8020 - val_loss: 0.9301\n",
      "Epoch 88/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9414 - loss: 0.1977 - val_accuracy: 0.8200 - val_loss: 0.7634\n",
      "Epoch 89/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.9307 - loss: 0.2074 - val_accuracy: 0.8340 - val_loss: 0.5823\n",
      "Epoch 90/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.9429 - loss: 0.1462 - val_accuracy: 0.8320 - val_loss: 0.8563\n",
      "Epoch 91/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9374 - loss: 0.1743 - val_accuracy: 0.8530 - val_loss: 0.6579\n",
      "Epoch 92/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step - accuracy: 0.9418 - loss: 0.1771 - val_accuracy: 0.8410 - val_loss: 0.8412\n",
      "Epoch 93/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9467 - loss: 0.1544 - val_accuracy: 0.8260 - val_loss: 0.8593\n",
      "Epoch 94/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9451 - loss: 0.1561 - val_accuracy: 0.8140 - val_loss: 0.9181\n",
      "Epoch 95/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9391 - loss: 0.1651 - val_accuracy: 0.8240 - val_loss: 0.7814\n",
      "Epoch 96/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9395 - loss: 0.2084 - val_accuracy: 0.8060 - val_loss: 0.8989\n",
      "Epoch 97/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9462 - loss: 0.1667 - val_accuracy: 0.8370 - val_loss: 0.7988\n",
      "Epoch 98/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9383 - loss: 0.1968 - val_accuracy: 0.8440 - val_loss: 0.6830\n",
      "Epoch 99/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - accuracy: 0.9441 - loss: 0.2020 - val_accuracy: 0.8420 - val_loss: 0.7575\n",
      "Epoch 100/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - accuracy: 0.9369 - loss: 0.2171 - val_accuracy: 0.8000 - val_loss: 1.1309\n"
     ]
    }
   ],
   "source": [
    "# training the model with data augmentation\n",
    "convnet_aug_history = convnet_model.fit(\n",
    "    train_dataset,\n",
    "    epochs = 100,\n",
    "    validation_data = validation_dataset,\n",
    "    callbacks = checkpoint_save(\"convnet_from_scratch_augmented.keras\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the training history\n",
    "plot_history(convnet_aug_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 8ms/step - accuracy: 0.8306 - loss: 0.4799\n",
      "Test accuracy: 0.836\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"convnet_from_scratch_augmented\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"convnet_from_scratch_augmented\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ sequential (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">178</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)   │           <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">89</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">87</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">43</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">41</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">9</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)      │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12544</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12544</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │        <span style=\"color: #00af00; text-decoration-color: #00af00\">12,545</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_3 (\u001b[38;5;33mCast\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ sequential (\u001b[38;5;33mSequential\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ rescaling_1 (\u001b[38;5;33mRescaling\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m178\u001b[0m, \u001b[38;5;34m32\u001b[0m)   │           \u001b[38;5;34m896\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m89\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_6 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m87\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m18,496\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m43\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_7 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m41\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │        \u001b[38;5;34m73,856\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_6 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m20\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_8 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m295,168\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_7 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m9\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_9 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m256\u001b[0m)      │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12544\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12544\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │        \u001b[38;5;34m12,545\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,982,088</span> (7.56 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,982,088\u001b[0m (7.56 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,041</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m991,041\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">991,047</span> (3.78 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m991,047\u001b[0m (3.78 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# evaluating the model on the test set\n",
    "convnet_aug_best = keras.models.load_model(\"convnet_from_scratch_augmented.keras\")\n",
    "convnet_aug_best_loss, convnet_aug_best_acc = convnet_aug_best.evaluate(test_dataset)\n",
    "print(f\"Test accuracy: {convnet_aug_best_acc:.3f}\")\n",
    "\n",
    "convnet_aug_best.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using Advanced Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-09 18:04:38.878371: E tensorflow/core/util/util.cc:131] oneDNN supports DT_HALF only on platforms with AVX-512. Falling back to the default Eigen-based implementation if present.\n",
      "I0000 00:00:1736413479.390404  184738 cuda_dnn.cc:529] Loaded cuDNN version 90600\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 327ms/step - accuracy: 0.5178 - loss: 0.9190 - val_accuracy: 0.5100 - val_loss: 0.6929\n",
      "Epoch 2/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 163ms/step - accuracy: 0.5867 - loss: 0.6857 - val_accuracy: 0.5000 - val_loss: 0.6945\n",
      "Epoch 3/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 187ms/step - accuracy: 0.6092 - loss: 0.6497 - val_accuracy: 0.5000 - val_loss: 0.7065\n",
      "Epoch 4/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 139ms/step - accuracy: 0.6110 - loss: 0.6506 - val_accuracy: 0.5000 - val_loss: 0.7186\n",
      "Epoch 5/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 113ms/step - accuracy: 0.6443 - loss: 0.6328 - val_accuracy: 0.5000 - val_loss: 0.7511\n",
      "Epoch 6/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 89ms/step - accuracy: 0.6559 - loss: 0.6157 - val_accuracy: 0.5270 - val_loss: 0.6975\n",
      "Epoch 7/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 95ms/step - accuracy: 0.6794 - loss: 0.6054 - val_accuracy: 0.5140 - val_loss: 0.7590\n",
      "Epoch 8/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 91ms/step - accuracy: 0.6863 - loss: 0.5934 - val_accuracy: 0.5000 - val_loss: 1.0494\n",
      "Epoch 9/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 97ms/step - accuracy: 0.7113 - loss: 0.5723 - val_accuracy: 0.5990 - val_loss: 0.6750\n",
      "Epoch 10/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 97ms/step - accuracy: 0.7294 - loss: 0.5350 - val_accuracy: 0.7300 - val_loss: 0.5236\n",
      "Epoch 11/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 111ms/step - accuracy: 0.7147 - loss: 0.5523 - val_accuracy: 0.7240 - val_loss: 0.5440\n",
      "Epoch 12/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 90ms/step - accuracy: 0.7563 - loss: 0.5163 - val_accuracy: 0.6280 - val_loss: 0.7511\n",
      "Epoch 13/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 105ms/step - accuracy: 0.7433 - loss: 0.5303 - val_accuracy: 0.7310 - val_loss: 0.5330\n",
      "Epoch 14/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 85ms/step - accuracy: 0.7605 - loss: 0.4980 - val_accuracy: 0.6830 - val_loss: 0.6950\n",
      "Epoch 15/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 178ms/step - accuracy: 0.8023 - loss: 0.4464 - val_accuracy: 0.6550 - val_loss: 0.8729\n",
      "Epoch 16/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 101ms/step - accuracy: 0.7776 - loss: 0.4688 - val_accuracy: 0.7650 - val_loss: 0.5022\n",
      "Epoch 17/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 118ms/step - accuracy: 0.7854 - loss: 0.4534 - val_accuracy: 0.6950 - val_loss: 0.6508\n",
      "Epoch 18/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 154ms/step - accuracy: 0.8013 - loss: 0.4343 - val_accuracy: 0.7600 - val_loss: 0.5196\n",
      "Epoch 19/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 151ms/step - accuracy: 0.8077 - loss: 0.4112 - val_accuracy: 0.7500 - val_loss: 0.5780\n",
      "Epoch 20/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 138ms/step - accuracy: 0.8191 - loss: 0.3913 - val_accuracy: 0.7860 - val_loss: 0.4565\n",
      "Epoch 21/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 120ms/step - accuracy: 0.8186 - loss: 0.4042 - val_accuracy: 0.7770 - val_loss: 0.4544\n",
      "Epoch 22/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 108ms/step - accuracy: 0.8304 - loss: 0.3861 - val_accuracy: 0.6750 - val_loss: 0.7142\n",
      "Epoch 23/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 127ms/step - accuracy: 0.8191 - loss: 0.4004 - val_accuracy: 0.7110 - val_loss: 0.7047\n",
      "Epoch 24/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 107ms/step - accuracy: 0.8250 - loss: 0.3731 - val_accuracy: 0.7450 - val_loss: 0.7218\n",
      "Epoch 25/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 129ms/step - accuracy: 0.8304 - loss: 0.3755 - val_accuracy: 0.7740 - val_loss: 0.4920\n",
      "Epoch 26/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 107ms/step - accuracy: 0.8354 - loss: 0.3718 - val_accuracy: 0.8210 - val_loss: 0.4177\n",
      "Epoch 27/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 105ms/step - accuracy: 0.8633 - loss: 0.3163 - val_accuracy: 0.7670 - val_loss: 0.4940\n",
      "Epoch 28/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 112ms/step - accuracy: 0.8452 - loss: 0.3619 - val_accuracy: 0.7880 - val_loss: 0.5088\n",
      "Epoch 29/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 108ms/step - accuracy: 0.8558 - loss: 0.3348 - val_accuracy: 0.8200 - val_loss: 0.4066\n",
      "Epoch 30/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.8669 - loss: 0.3068 - val_accuracy: 0.7300 - val_loss: 0.6931\n",
      "Epoch 31/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 94ms/step - accuracy: 0.8700 - loss: 0.2945 - val_accuracy: 0.6420 - val_loss: 0.7396\n",
      "Epoch 32/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 95ms/step - accuracy: 0.8615 - loss: 0.3223 - val_accuracy: 0.7550 - val_loss: 0.6001\n",
      "Epoch 33/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 107ms/step - accuracy: 0.8669 - loss: 0.3068 - val_accuracy: 0.7970 - val_loss: 0.4763\n",
      "Epoch 34/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 128ms/step - accuracy: 0.8617 - loss: 0.3030 - val_accuracy: 0.7810 - val_loss: 0.5506\n",
      "Epoch 35/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 112ms/step - accuracy: 0.8793 - loss: 0.2872 - val_accuracy: 0.8200 - val_loss: 0.5150\n",
      "Epoch 36/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 122ms/step - accuracy: 0.8809 - loss: 0.2888 - val_accuracy: 0.7730 - val_loss: 0.5673\n",
      "Epoch 37/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 217ms/step - accuracy: 0.8820 - loss: 0.2760 - val_accuracy: 0.8240 - val_loss: 0.3851\n",
      "Epoch 38/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 147ms/step - accuracy: 0.8952 - loss: 0.2622 - val_accuracy: 0.7540 - val_loss: 0.9465\n",
      "Epoch 39/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 162ms/step - accuracy: 0.9007 - loss: 0.2503 - val_accuracy: 0.8220 - val_loss: 0.3924\n",
      "Epoch 40/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 105ms/step - accuracy: 0.8859 - loss: 0.2581 - val_accuracy: 0.8510 - val_loss: 0.3878\n",
      "Epoch 41/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 102ms/step - accuracy: 0.8973 - loss: 0.2520 - val_accuracy: 0.8530 - val_loss: 0.3377\n",
      "Epoch 42/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 120ms/step - accuracy: 0.8826 - loss: 0.2751 - val_accuracy: 0.7620 - val_loss: 0.8568\n",
      "Epoch 43/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 93ms/step - accuracy: 0.9096 - loss: 0.2340 - val_accuracy: 0.8280 - val_loss: 0.4938\n",
      "Epoch 44/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 147ms/step - accuracy: 0.8993 - loss: 0.2349 - val_accuracy: 0.8430 - val_loss: 0.4136\n",
      "Epoch 45/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9273 - loss: 0.1927 - val_accuracy: 0.7200 - val_loss: 0.7930\n",
      "Epoch 46/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9017 - loss: 0.2411 - val_accuracy: 0.8310 - val_loss: 0.4788\n",
      "Epoch 47/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 113ms/step - accuracy: 0.9069 - loss: 0.2391 - val_accuracy: 0.8620 - val_loss: 0.3330\n",
      "Epoch 48/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 117ms/step - accuracy: 0.9065 - loss: 0.2542 - val_accuracy: 0.8830 - val_loss: 0.3061\n",
      "Epoch 49/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 127ms/step - accuracy: 0.9310 - loss: 0.1937 - val_accuracy: 0.8250 - val_loss: 0.4139\n",
      "Epoch 50/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 145ms/step - accuracy: 0.9130 - loss: 0.2170 - val_accuracy: 0.8520 - val_loss: 0.3545\n",
      "Epoch 51/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 107ms/step - accuracy: 0.9039 - loss: 0.2584 - val_accuracy: 0.7290 - val_loss: 1.2404\n",
      "Epoch 52/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 119ms/step - accuracy: 0.9271 - loss: 0.1780 - val_accuracy: 0.8850 - val_loss: 0.2826\n",
      "Epoch 53/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 109ms/step - accuracy: 0.9211 - loss: 0.2103 - val_accuracy: 0.7960 - val_loss: 0.5422\n",
      "Epoch 54/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 124ms/step - accuracy: 0.9295 - loss: 0.1750 - val_accuracy: 0.6250 - val_loss: 2.0180\n",
      "Epoch 55/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 97ms/step - accuracy: 0.9265 - loss: 0.1863 - val_accuracy: 0.8640 - val_loss: 0.4433\n",
      "Epoch 56/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 105ms/step - accuracy: 0.9222 - loss: 0.1842 - val_accuracy: 0.8720 - val_loss: 0.3124\n",
      "Epoch 57/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 119ms/step - accuracy: 0.9083 - loss: 0.2250 - val_accuracy: 0.8540 - val_loss: 0.3413\n",
      "Epoch 58/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 114ms/step - accuracy: 0.9092 - loss: 0.2095 - val_accuracy: 0.7410 - val_loss: 1.5213\n",
      "Epoch 59/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 101ms/step - accuracy: 0.9335 - loss: 0.1717 - val_accuracy: 0.8860 - val_loss: 0.2844\n",
      "Epoch 60/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 87ms/step - accuracy: 0.9220 - loss: 0.1943 - val_accuracy: 0.8100 - val_loss: 0.6428\n",
      "Epoch 61/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9443 - loss: 0.1628 - val_accuracy: 0.7590 - val_loss: 1.1382\n",
      "Epoch 62/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9327 - loss: 0.1574 - val_accuracy: 0.8510 - val_loss: 0.4473\n",
      "Epoch 63/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 100ms/step - accuracy: 0.9422 - loss: 0.1404 - val_accuracy: 0.7720 - val_loss: 0.9956\n",
      "Epoch 64/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 91ms/step - accuracy: 0.9324 - loss: 0.1816 - val_accuracy: 0.7600 - val_loss: 0.9591\n",
      "Epoch 65/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 93ms/step - accuracy: 0.9471 - loss: 0.1346 - val_accuracy: 0.8730 - val_loss: 0.3066\n",
      "Epoch 66/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 102ms/step - accuracy: 0.9329 - loss: 0.1741 - val_accuracy: 0.7580 - val_loss: 1.0224\n",
      "Epoch 67/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 102ms/step - accuracy: 0.9414 - loss: 0.1370 - val_accuracy: 0.8380 - val_loss: 0.5678\n",
      "Epoch 68/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 91ms/step - accuracy: 0.9313 - loss: 0.1610 - val_accuracy: 0.8520 - val_loss: 0.4189\n",
      "Epoch 69/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 92ms/step - accuracy: 0.9358 - loss: 0.1760 - val_accuracy: 0.7830 - val_loss: 0.8156\n",
      "Epoch 70/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 80ms/step - accuracy: 0.9318 - loss: 0.1633 - val_accuracy: 0.8450 - val_loss: 0.4326\n",
      "Epoch 71/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 91ms/step - accuracy: 0.9633 - loss: 0.1150 - val_accuracy: 0.7990 - val_loss: 0.7789\n",
      "Epoch 72/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 109ms/step - accuracy: 0.9399 - loss: 0.1348 - val_accuracy: 0.8870 - val_loss: 0.3119\n",
      "Epoch 73/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 107ms/step - accuracy: 0.9449 - loss: 0.1292 - val_accuracy: 0.7950 - val_loss: 0.5420\n",
      "Epoch 74/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 105ms/step - accuracy: 0.9351 - loss: 0.1560 - val_accuracy: 0.8320 - val_loss: 0.5105\n",
      "Epoch 75/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 109ms/step - accuracy: 0.9471 - loss: 0.1436 - val_accuracy: 0.8950 - val_loss: 0.2972\n",
      "Epoch 76/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 104ms/step - accuracy: 0.9367 - loss: 0.1542 - val_accuracy: 0.8730 - val_loss: 0.3203\n",
      "Epoch 77/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 121ms/step - accuracy: 0.9493 - loss: 0.1208 - val_accuracy: 0.8690 - val_loss: 0.3579\n",
      "Epoch 78/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9536 - loss: 0.1126 - val_accuracy: 0.8370 - val_loss: 0.6029\n",
      "Epoch 79/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9537 - loss: 0.1212 - val_accuracy: 0.9020 - val_loss: 0.2764\n",
      "Epoch 80/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 85ms/step - accuracy: 0.9485 - loss: 0.1180 - val_accuracy: 0.9020 - val_loss: 0.2555\n",
      "Epoch 81/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9534 - loss: 0.1244 - val_accuracy: 0.8360 - val_loss: 0.4293\n",
      "Epoch 82/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9514 - loss: 0.1931 - val_accuracy: 0.8170 - val_loss: 0.8994\n",
      "Epoch 83/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9529 - loss: 0.1309 - val_accuracy: 0.8430 - val_loss: 0.6390\n",
      "Epoch 84/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9417 - loss: 0.1330 - val_accuracy: 0.7760 - val_loss: 1.1863\n",
      "Epoch 85/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 83ms/step - accuracy: 0.9527 - loss: 0.1142 - val_accuracy: 0.8190 - val_loss: 0.4663\n",
      "Epoch 86/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 99ms/step - accuracy: 0.9526 - loss: 0.1382 - val_accuracy: 0.8210 - val_loss: 0.5657\n",
      "Epoch 87/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 100ms/step - accuracy: 0.9596 - loss: 0.1182 - val_accuracy: 0.8510 - val_loss: 0.5457\n",
      "Epoch 88/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 97ms/step - accuracy: 0.9597 - loss: 0.0949 - val_accuracy: 0.9010 - val_loss: 0.3396\n",
      "Epoch 89/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 103ms/step - accuracy: 0.9599 - loss: 0.0914 - val_accuracy: 0.8440 - val_loss: 0.5053\n",
      "Epoch 90/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 96ms/step - accuracy: 0.9483 - loss: 0.1360 - val_accuracy: 0.8040 - val_loss: 0.7024\n",
      "Epoch 91/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9710 - loss: 0.0930 - val_accuracy: 0.8240 - val_loss: 0.7123\n",
      "Epoch 92/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 96ms/step - accuracy: 0.9599 - loss: 0.0949 - val_accuracy: 0.7940 - val_loss: 0.8262\n",
      "Epoch 93/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 96ms/step - accuracy: 0.9621 - loss: 0.1107 - val_accuracy: 0.8720 - val_loss: 0.4045\n",
      "Epoch 94/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 95ms/step - accuracy: 0.9452 - loss: 0.1344 - val_accuracy: 0.7950 - val_loss: 1.0577\n",
      "Epoch 95/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 92ms/step - accuracy: 0.9559 - loss: 0.1285 - val_accuracy: 0.8580 - val_loss: 0.5861\n",
      "Epoch 96/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 97ms/step - accuracy: 0.9584 - loss: 0.1078 - val_accuracy: 0.7650 - val_loss: 1.3844\n",
      "Epoch 97/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9735 - loss: 0.0771 - val_accuracy: 0.8910 - val_loss: 0.4525\n",
      "Epoch 98/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 94ms/step - accuracy: 0.9601 - loss: 0.1368 - val_accuracy: 0.8120 - val_loss: 0.8481\n",
      "Epoch 99/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9584 - loss: 0.1038 - val_accuracy: 0.9030 - val_loss: 0.2721\n",
      "Epoch 100/100\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 98ms/step - accuracy: 0.9559 - loss: 0.0912 - val_accuracy: 0.8460 - val_loss: 0.5243\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "inputs = keras.Input(shape=(180,180,3))\n",
    "x = data_augmentation(inputs)\n",
    "x = layers.Rescaling(1./255)(x)\n",
    "x = layers.Conv2D(filters=32, kernel_size=5, use_bias=False)(x)\n",
    "\n",
    "for size in [32, 64, 128, 256, 512]:\n",
    "\tresidual = x\n",
    "\t\n",
    "\tx = layers.BatchNormalization()(x)\n",
    "\tx = layers.Activation(\"relu\")(x)\n",
    "\tx = layers.SeparableConv2D(size, 3, padding=\"same\", use_bias=False)(x)\n",
    "\t\n",
    "\tx = layers.BatchNormalization()(x)\n",
    "\tx = layers.Activation(\"relu\")(x)\n",
    "\tx = layers.SeparableConv2D(size, 3, padding=\"same\", use_bias=False)(x)\n",
    "\t\n",
    "\tx = layers.MaxPooling2D(3, strides=2, padding=\"same\")(x)\n",
    "\t\n",
    "\tresidual = layers.Conv2D(size, 1, strides=2, padding=\"same\", use_bias=False)(residual)\n",
    "\tx = layers.add([x, residual])\n",
    "\n",
    "x = layers.GlobalAveragePooling2D()(x)\n",
    "x = layers.Dropout(0.5)(x)\n",
    "outputs = layers.Dense(1, activation=\"sigmoid\")(x)\n",
    "convnet_adv_architecture = keras.Model(inputs=inputs, outputs=outputs)\n",
    "\n",
    "convnet_adv_architecture.compile(optimizer='rmsprop',\n",
    "\t\t\t\t\t\t\t\tloss='binary_crossentropy',\n",
    "\t\t\t\t\t\t\t\tmetrics=['accuracy'])\n",
    "\n",
    "convnet_adv_history = convnet_adv_architecture.fit(\n",
    "    train_dataset,\n",
    "    epochs = 100,\n",
    "    validation_data = validation_dataset,\n",
    "    callbacks = checkpoint_save(\"convnet_from_scratch_augmented.keras\")\n",
    ")\n",
    "\n",
    "plot_history(convnet_adv_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 31ms/step - accuracy: 0.8970 - loss: 0.3343\n",
      "Test accuracy: 0.891\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_3\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"functional_3\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)        </span>┃<span style=\"font-weight: bold\"> Output Shape      </span>┃<span style=\"font-weight: bold\">    Param # </span>┃<span style=\"font-weight: bold\"> Connected to      </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_4       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ input_layer_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ cast_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ rescaling_2         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_12 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,400</span> │ rescaling_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ conv2d_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_20       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_20 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,312</span> │ activation_20[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_21       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_21 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">176</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,312</span> │ activation_21[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_10    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_13 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_10 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_10… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │ conv2d_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ add_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_22       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_22 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,336</span> │ activation_22[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_23       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_23 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">88</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,672</span> │ activation_23[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_11    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_14 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ add_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_11 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_11… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │ conv2d_14[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │ add_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_24       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_24 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">8,768</span> │ activation_24[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_25       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_25 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">44</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">17,536</span> │ activation_25[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_12    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_15 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">8,192</span> │ add_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_12 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_12… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │ conv2d_15[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ add_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_26       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_26 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">33,920</span> │ activation_26[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_27       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_27 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">67,840</span> │ activation_27[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_13    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">32,768</span> │ add_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_13 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_13… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │ conv2d_16[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ add_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_28       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_28 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">133,376</span> │ activation_28[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_29       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_29 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">266,752</span> │ activation_29[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_14    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ separable_conv2d… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_17 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">131,072</span> │ add_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_14 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_14… │\n",
       "│                     │                   │            │ conv2d_17[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ global_average_poo… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add_14[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ global_average_p… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)         │        <span style=\"color: #00af00; text-decoration-color: #00af00\">513</span> │ dropout_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)       \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m   Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to     \u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_4       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
       "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_3 (\u001b[38;5;33mCast\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ input_layer_4[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ cast_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mSequential\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ rescaling_2         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mRescaling\u001b[0m)         │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_12 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │      \u001b[38;5;34m2,400\u001b[0m │ rescaling_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
       "│                     │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │        \u001b[38;5;34m128\u001b[0m │ conv2d_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_20       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_20 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │      \u001b[38;5;34m1,312\u001b[0m │ activation_20[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │        \u001b[38;5;34m128\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_21       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_21 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m176\u001b[0m, \u001b[38;5;34m176\u001b[0m,  │      \u001b[38;5;34m1,312\u001b[0m │ activation_21[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_10    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_13 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_10 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_10… │\n",
       "│                     │ \u001b[38;5;34m32\u001b[0m)               │            │ conv2d_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │        \u001b[38;5;34m128\u001b[0m │ add_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_22       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_22 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │      \u001b[38;5;34m2,336\u001b[0m │ activation_22[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_23       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_23 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m88\u001b[0m, \u001b[38;5;34m88\u001b[0m,    │      \u001b[38;5;34m4,672\u001b[0m │ activation_23[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_11    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_14 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │      \u001b[38;5;34m2,048\u001b[0m │ add_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_11 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_11… │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │ conv2d_14[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ add_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_24       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_24 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │      \u001b[38;5;34m8,768\u001b[0m │ activation_24[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │        \u001b[38;5;34m512\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_25       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_25 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m44\u001b[0m, \u001b[38;5;34m44\u001b[0m,    │     \u001b[38;5;34m17,536\u001b[0m │ activation_25[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_12    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_15 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │      \u001b[38;5;34m8,192\u001b[0m │ add_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_12 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_12… │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │ conv2d_15[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │        \u001b[38;5;34m512\u001b[0m │ add_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_26       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_26 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │     \u001b[38;5;34m33,920\u001b[0m │ activation_26[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_27       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_27 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m,    │     \u001b[38;5;34m67,840\u001b[0m │ activation_27[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_13    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_16 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │     \u001b[38;5;34m32,768\u001b[0m │ add_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_13 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_13… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │ conv2d_16[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ add_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_28       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_28 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │    \u001b[38;5;34m133,376\u001b[0m │ activation_28[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │      \u001b[38;5;34m2,048\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_29       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ separable_conv2d_29 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m,    │    \u001b[38;5;34m266,752\u001b[0m │ activation_29[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_14    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ separable_conv2d… │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_17 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m512\u001b[0m) │    \u001b[38;5;34m131,072\u001b[0m │ add_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_14 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_14… │\n",
       "│                     │                   │            │ conv2d_17[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ global_average_poo… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ add_14[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ global_average_p… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)         │        \u001b[38;5;34m513\u001b[0m │ dropout_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">1,440,712</span> (5.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,440,712\u001b[0m (5.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">718,849</span> (2.74 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m718,849\u001b[0m (2.74 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,008</span> (11.75 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m3,008\u001b[0m (11.75 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">718,855</span> (2.74 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m718,855\u001b[0m (2.74 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# evaluating the model on the test set\n",
    "convnet_adv_best = keras.models.load_model(\"convnet_from_scratch_augmented.keras\")\n",
    "convnet_adv_best_loss, convnet_adv_best_acc = convnet_adv_best.evaluate(test_dataset)\n",
    "print(f\"Test accuracy: {convnet_adv_best_acc:.3f}\")\n",
    "\n",
    "convnet_adv_best.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using a Pre-Trained Model (VGG16)\n",
    "### Feature-Extraction using the Pre-Trained Model\n",
    "#### Loading the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"vgg16\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"vgg16\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,792</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">1,180,160</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_3 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_4 (\u001b[38;5;33mCast\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │         \u001b[38;5;34m1,792\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │        \u001b[38;5;34m36,928\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │        \u001b[38;5;34m73,856\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │       \u001b[38;5;34m147,584\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m295,168\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m1,180,160\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# loading the pre-trained base of VGG16\n",
    "vgg16_base = keras.applications.VGG16(\n",
    "    weights = 'imagenet',\n",
    "    include_top = False,\n",
    "    input_shape = (180, 180, 3)\n",
    ")\n",
    "\n",
    "vgg16_base.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Extracting Features amd Labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-08 18:34:40.913690: I tensorflow/core/framework/local_rendezvous.cc:405] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "2025-01-08 18:34:43.869574: I tensorflow/core/framework/local_rendezvous.cc:405] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "((2000, 5, 5, 512), (1000, 5, 5, 512), (2000, 5, 5, 512))"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extracting features from the dataset using the pre-trained VGG16 base\n",
    "def get_feature_and_labels(dataset):\n",
    "    all_features = []\n",
    "    all_labels = []\n",
    "\n",
    "    for images, labels in dataset:\n",
    "        preprocessed_images = keras.applications.vgg16.preprocess_input(images)\n",
    "        features = vgg16_base(preprocessed_images)\n",
    "        all_features.append(features)\n",
    "        all_labels.append(labels)\n",
    "\n",
    "    return np.concatenate(all_features), np.concatenate(all_labels)\n",
    "\n",
    "train_features, train_labels = get_feature_and_labels(train_dataset)\n",
    "val_features, val_labels = get_feature_and_labels(validation_dataset)\n",
    "test_features, test_labels = get_feature_and_labels(test_dataset)\n",
    "\n",
    "train_features.shape, val_features.shape, test_features.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training Top(=MLP) Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "vgg16_top_input = keras.Input(shape=(5, 5, 512))\n",
    "\n",
    "vgg16_top_layer = layers.Flatten()(vgg16_top_input)\n",
    "vgg16_top_layer = layers.Dense(256, activation='relu')(vgg16_top_layer)\n",
    "vgg16_top_layer = layers.Dropout(0.5)(vgg16_top_layer)\n",
    "\n",
    "vgg16_top_output = layers.Dense(1, activation='sigmoid')(vgg16_top_layer)\n",
    "vgg16_top = keras.Model(inputs = vgg16_top_input, outputs = vgg16_top_output, name=\"vgg16_top\")\n",
    "\n",
    "vgg16_top.compile(optimizer='rmsprop',\n",
    "            loss='binary_crossentropy',\n",
    "            metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 16ms/step - accuracy: 0.7705 - loss: 3.3115"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-08 18:34:47.164288: I external/local_xla/xla/stream_executor/cuda/cuda_asm_compiler.cc:397] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_26', 4 bytes spill stores, 4 bytes spill loads\n",
      "\n",
      "2025-01-08 18:34:47.259709: I external/local_xla/xla/stream_executor/cuda/cuda_asm_compiler.cc:397] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_26', 52 bytes spill stores, 52 bytes spill loads\n",
      "\n",
      "2025-01-08 18:34:47.819848: I external/local_xla/xla/stream_executor/cuda/cuda_asm_compiler.cc:397] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_26', 4 bytes spill stores, 4 bytes spill loads\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 41ms/step - accuracy: 0.7718 - loss: 3.2929 - val_accuracy: 0.9570 - val_loss: 0.6406\n",
      "Epoch 2/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - accuracy: 0.9093 - loss: 1.3890 - val_accuracy: 0.9580 - val_loss: 0.5941\n",
      "Epoch 3/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9334 - loss: 1.0461 - val_accuracy: 0.9240 - val_loss: 1.1376\n",
      "Epoch 4/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.9296 - loss: 1.0559 - val_accuracy: 0.9700 - val_loss: 0.4672\n",
      "Epoch 5/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9573 - loss: 0.6407 - val_accuracy: 0.9270 - val_loss: 1.1134\n",
      "Epoch 6/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9451 - loss: 0.8295 - val_accuracy: 0.9660 - val_loss: 0.5229\n",
      "Epoch 7/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.9612 - loss: 0.6039 - val_accuracy: 0.9710 - val_loss: 0.4530\n",
      "Epoch 8/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9554 - loss: 0.6770 - val_accuracy: 0.9710 - val_loss: 0.4637\n",
      "Epoch 9/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9576 - loss: 0.6559 - val_accuracy: 0.9750 - val_loss: 0.3865\n",
      "Epoch 10/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9589 - loss: 0.6487 - val_accuracy: 0.9670 - val_loss: 0.5083\n",
      "Epoch 11/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.9607 - loss: 0.6245 - val_accuracy: 0.9760 - val_loss: 0.3545\n",
      "Epoch 12/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9557 - loss: 0.6808 - val_accuracy: 0.9550 - val_loss: 0.6964\n",
      "Epoch 13/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9569 - loss: 0.6790 - val_accuracy: 0.9670 - val_loss: 0.5172\n",
      "Epoch 14/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9647 - loss: 0.5484 - val_accuracy: 0.9680 - val_loss: 0.4919\n",
      "Epoch 15/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9664 - loss: 0.5108 - val_accuracy: 0.9710 - val_loss: 0.4206\n",
      "Epoch 16/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9598 - loss: 0.6132 - val_accuracy: 0.9720 - val_loss: 0.4052\n",
      "Epoch 17/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9665 - loss: 0.5335 - val_accuracy: 0.9610 - val_loss: 0.5695\n",
      "Epoch 18/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9598 - loss: 0.6269 - val_accuracy: 0.9680 - val_loss: 0.5019\n",
      "Epoch 19/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9636 - loss: 0.5761 - val_accuracy: 0.9650 - val_loss: 0.5444\n",
      "Epoch 20/20\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9591 - loss: 0.6288 - val_accuracy: 0.9740 - val_loss: 0.4073\n"
     ]
    }
   ],
   "source": [
    "vgg16_top_history = vgg16_top.fit(\n",
    "    train_features, train_labels,\n",
    "    epochs = 20,\n",
    "    validation_data = (val_features, val_labels),\n",
    "    callbacks = checkpoint_save(\"vgg16_top.keras\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualising Model-Training history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m60/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.9676 - loss: 0.5112"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-01-08 18:34:55.477807: I external/local_xla/xla/stream_executor/cuda/cuda_asm_compiler.cc:397] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_26', 4 bytes spill stores, 4 bytes spill loads\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 11ms/step - accuracy: 0.9678 - loss: 0.5081\n",
      "Test accuracy: 0.971\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"vgg16_top\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"vgg16_top\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12800</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)            │     <span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,056</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">257</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_4 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_6 (\u001b[38;5;33mCast\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_2 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12800\u001b[0m)          │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)            │     \u001b[38;5;34m3,277,056\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │           \u001b[38;5;34m257\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">6,554,632</span> (25.00 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m6,554,632\u001b[0m (25.00 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,313</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3,277,313\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,319</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m3,277,319\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the training history\n",
    "plot_history(vgg16_top_history)\n",
    "\n",
    "vgg16_top_best = keras.models.load_model(\"vgg16_top.keras\")\n",
    "vgg16_loss, vgg16_acc = vgg16_top_best.evaluate(test_features, test_labels)\n",
    "print(f\"Test accuracy: {vgg16_acc:.3f}\")\n",
    "\n",
    "vgg16_top_best.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Buildign the End-To-End Model on a Pre-Trained Model\n",
    "#### Reloading and Freezing the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"vgg16\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"vgg16\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,792</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">90</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">45</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │       <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">1,180,160</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">22</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_pool (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_5 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ cast_7 (\u001b[38;5;33mCast\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │         \u001b[38;5;34m1,792\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │        \u001b[38;5;34m36,928\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block1_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │        \u001b[38;5;34m73,856\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m90\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │       \u001b[38;5;34m147,584\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block2_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m128\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m295,168\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m45\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │       \u001b[38;5;34m590,080\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block3_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m256\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m1,180,160\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m22\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block4_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv1 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv2 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_conv3 (\u001b[38;5;33mConv2D\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m512\u001b[0m)    │     \u001b[38;5;34m2,359,808\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ block5_pool (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# freezing pre-trained VGG16 base to train the top layers\n",
    "vgg16_base_freeze = keras.applications.VGG16(\n",
    "    weights = 'imagenet',\n",
    "    include_top = False,\n",
    "    input_shape = (180, 180, 3)\n",
    ")\n",
    "\n",
    "vgg16_base_freeze.trainable = False\n",
    "vgg16_base_freeze.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Building the End-To-End Model with Data Augmentation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# designing and compiling the model with the frozen pre-trained VGG16 base\n",
    "vgg16_freeze_input = keras.Input(shape=(180, 180, 3))\n",
    "\n",
    "vgg16_freeze_layers = data_augmentation(vgg16_freeze_input)\n",
    "vgg16_freeze_layers = keras.applications.vgg16.preprocess_input(vgg16_freeze_layers)\n",
    "vgg16_freeze_layers = vgg16_base_freeze(vgg16_freeze_layers)\n",
    "vgg16_freeze_layers = layers.Flatten()(vgg16_freeze_layers)\n",
    "vgg16_freeze_layers = layers.Dense(256, activation='relu')(vgg16_freeze_layers)\n",
    "vgg16_freeze_layers = layers.Dropout(0.5)(vgg16_freeze_layers)\n",
    "\n",
    "vgg16_freeze_output = layers.Dense(1, activation='sigmoid')(vgg16_freeze_layers)\n",
    "vgg16_freeze_model = keras.Model(inputs = vgg16_freeze_input, outputs = vgg16_freeze_output, name=\"vgg16_freeze\")\n",
    "vgg16_freeze_model.compile(optimizer='rmsprop',\n",
    "                            loss='binary_crossentropy',\n",
    "                            metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 50ms/step - accuracy: 0.4914 - loss: 8.0113 - val_accuracy: 0.5000 - val_loss: 8.0590\n",
      "Epoch 2/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 48ms/step - accuracy: 0.5265 - loss: 7.5860 - val_accuracy: 0.8060 - val_loss: 3.0299\n",
      "Epoch 3/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 45ms/step - accuracy: 0.8205 - loss: 2.7752 - val_accuracy: 0.9530 - val_loss: 0.7419\n",
      "Epoch 4/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - accuracy: 0.9008 - loss: 1.5077 - val_accuracy: 0.9160 - val_loss: 1.2669\n",
      "Epoch 5/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 44ms/step - accuracy: 0.8851 - loss: 1.7795 - val_accuracy: 0.9620 - val_loss: 0.5898\n",
      "Epoch 6/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 43ms/step - accuracy: 0.9212 - loss: 1.1898 - val_accuracy: 0.9640 - val_loss: 0.5381\n",
      "Epoch 7/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9194 - loss: 1.2322 - val_accuracy: 0.9510 - val_loss: 0.7623\n",
      "Epoch 8/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.8877 - loss: 1.6944 - val_accuracy: 0.9620 - val_loss: 0.5830\n",
      "Epoch 9/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 46ms/step - accuracy: 0.9157 - loss: 1.3040 - val_accuracy: 0.9650 - val_loss: 0.5232\n",
      "Epoch 10/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 47ms/step - accuracy: 0.9081 - loss: 1.3946 - val_accuracy: 0.9680 - val_loss: 0.5041\n",
      "Epoch 11/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9333 - loss: 1.0403 - val_accuracy: 0.9610 - val_loss: 0.5902\n",
      "Epoch 12/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9257 - loss: 1.1662 - val_accuracy: 0.9650 - val_loss: 0.5138\n",
      "Epoch 13/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9353 - loss: 0.9821 - val_accuracy: 0.9760 - val_loss: 0.3728\n",
      "Epoch 14/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9376 - loss: 0.9428 - val_accuracy: 0.9660 - val_loss: 0.5298\n",
      "Epoch 15/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9482 - loss: 0.8104 - val_accuracy: 0.9660 - val_loss: 0.5140\n",
      "Epoch 16/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 44ms/step - accuracy: 0.9525 - loss: 0.7379 - val_accuracy: 0.9770 - val_loss: 0.3371\n",
      "Epoch 17/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9511 - loss: 0.7569 - val_accuracy: 0.9570 - val_loss: 0.6524\n",
      "Epoch 18/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9191 - loss: 1.2952 - val_accuracy: 0.9570 - val_loss: 0.6807\n",
      "Epoch 19/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9255 - loss: 1.1896 - val_accuracy: 0.9700 - val_loss: 0.4552\n",
      "Epoch 20/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9504 - loss: 0.7826 - val_accuracy: 0.9600 - val_loss: 0.6330\n",
      "Epoch 21/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9451 - loss: 0.8434 - val_accuracy: 0.9690 - val_loss: 0.4455\n",
      "Epoch 22/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9582 - loss: 0.6510 - val_accuracy: 0.9700 - val_loss: 0.4355\n",
      "Epoch 23/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9486 - loss: 0.7931 - val_accuracy: 0.9650 - val_loss: 0.5074\n",
      "Epoch 24/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9439 - loss: 0.8416 - val_accuracy: 0.9770 - val_loss: 0.3530\n",
      "Epoch 25/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 40ms/step - accuracy: 0.9508 - loss: 0.7525 - val_accuracy: 0.9300 - val_loss: 1.1129\n",
      "Epoch 26/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9019 - loss: 1.5625 - val_accuracy: 0.9780 - val_loss: 0.3308\n",
      "Epoch 27/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9516 - loss: 0.7558 - val_accuracy: 0.9670 - val_loss: 0.5260\n",
      "Epoch 28/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - accuracy: 0.9415 - loss: 0.9387 - val_accuracy: 0.9720 - val_loss: 0.4363\n",
      "Epoch 29/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9415 - loss: 0.8857 - val_accuracy: 0.9700 - val_loss: 0.4784\n",
      "Epoch 30/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9603 - loss: 0.6204 - val_accuracy: 0.9760 - val_loss: 0.3511\n",
      "Epoch 31/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9532 - loss: 0.7225 - val_accuracy: 0.9750 - val_loss: 0.3876\n",
      "Epoch 32/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9551 - loss: 0.6595 - val_accuracy: 0.9750 - val_loss: 0.4008\n",
      "Epoch 33/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9476 - loss: 0.8255 - val_accuracy: 0.9690 - val_loss: 0.4834\n",
      "Epoch 34/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9461 - loss: 0.8525 - val_accuracy: 0.9740 - val_loss: 0.4046\n",
      "Epoch 35/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9418 - loss: 0.9075 - val_accuracy: 0.9780 - val_loss: 0.3535\n",
      "Epoch 36/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9524 - loss: 0.7059 - val_accuracy: 0.9760 - val_loss: 0.3844\n",
      "Epoch 37/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9570 - loss: 0.6722 - val_accuracy: 0.9790 - val_loss: 0.3151\n",
      "Epoch 38/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9560 - loss: 0.6816 - val_accuracy: 0.9700 - val_loss: 0.4752\n",
      "Epoch 39/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 45ms/step - accuracy: 0.9433 - loss: 0.8707 - val_accuracy: 0.9630 - val_loss: 0.5869\n",
      "Epoch 40/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9427 - loss: 0.8849 - val_accuracy: 0.9740 - val_loss: 0.4071\n",
      "Epoch 41/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - accuracy: 0.9596 - loss: 0.6423 - val_accuracy: 0.9770 - val_loss: 0.3368\n",
      "Epoch 42/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9534 - loss: 0.7305 - val_accuracy: 0.9770 - val_loss: 0.3541\n",
      "Epoch 43/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9593 - loss: 0.6418 - val_accuracy: 0.9710 - val_loss: 0.4332\n",
      "Epoch 44/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 40ms/step - accuracy: 0.9512 - loss: 0.7633 - val_accuracy: 0.9790 - val_loss: 0.3221\n",
      "Epoch 45/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 40ms/step - accuracy: 0.9534 - loss: 0.7240 - val_accuracy: 0.9750 - val_loss: 0.3758\n",
      "Epoch 46/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9603 - loss: 0.6251 - val_accuracy: 0.9710 - val_loss: 0.4664\n",
      "Epoch 47/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 41ms/step - accuracy: 0.9492 - loss: 0.8013 - val_accuracy: 0.9690 - val_loss: 0.4990\n",
      "Epoch 48/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9456 - loss: 0.7860 - val_accuracy: 0.9680 - val_loss: 0.4918\n",
      "Epoch 49/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9537 - loss: 0.7146 - val_accuracy: 0.9780 - val_loss: 0.3530\n",
      "Epoch 50/50\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9650 - loss: 0.5482 - val_accuracy: 0.9800 - val_loss: 0.3213\n"
     ]
    }
   ],
   "source": [
    "# training the model with the frozen pre-trained VGG16 base\n",
    "vgg16_freeze_history = vgg16_freeze_model.fit(\n",
    "    train_dataset,\n",
    "    epochs = 50,\n",
    "    validation_data = validation_dataset,\n",
    "    callbacks = checkpoint_save(\"vgg16_freeze.keras\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dogs-vs-cats/\n",
    "dogs-vs-cats_small/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualizing Model-Training History"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9715 - loss: 0.4567\n",
      "Test accuracy: 0.972\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"vgg16_freeze\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"vgg16_freeze\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)        </span>┃<span style=\"font-weight: bold\"> Output Shape      </span>┃<span style=\"font-weight: bold\">    Param # </span>┃<span style=\"font-weight: bold\"> Connected to      </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_6       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ input_layer_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ cast_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_1          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_2          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ stack_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Stack</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ get_item[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],   │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │ get_item_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>], │\n",
       "│                     │                   │            │ get_item_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ stack_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ vgg16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Functional</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> │ add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]         │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ flatten_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12800</span>)     │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ vgg16[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │  <span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,056</span> │ flatten_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ dense_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)         │        <span style=\"color: #00af00; text-decoration-color: #00af00\">257</span> │ dropout_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)       \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m   Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to     \u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_6       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
       "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_9 (\u001b[38;5;33mCast\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ input_layer_6[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ cast_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mSequential\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item (\u001b[38;5;33mGetItem\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_1          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mGetItem\u001b[0m)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_2          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mGetItem\u001b[0m)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ stack_1 (\u001b[38;5;33mStack\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ get_item[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],   │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │ get_item_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
       "│                     │                   │            │ get_item_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (\u001b[38;5;33mAdd\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ stack_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ vgg16 (\u001b[38;5;33mFunctional\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m14,714,688\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]         │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ flatten_3 (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12800\u001b[0m)     │          \u001b[38;5;34m0\u001b[0m │ vgg16[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │  \u001b[38;5;34m3,277,056\u001b[0m │ flatten_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ dense_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_5 (\u001b[38;5;33mDense\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)         │        \u001b[38;5;34m257\u001b[0m │ dropout_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">21,269,320</span> (81.14 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m21,269,320\u001b[0m (81.14 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,313</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3,277,313\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,319</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m3,277,319\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the training history\n",
    "plot_history(vgg16_freeze_history)\n",
    "\n",
    "vgg16_freeze_best = keras.models.load_model(\"vgg16_freeze.keras\")\n",
    "vgg16_freeze_loss, vgg16_freeze_acc = vgg16_freeze_best.evaluate(test_dataset)\n",
    "print(f\"Test accuracy: {vgg16_freeze_acc:.3f}\")\n",
    "\n",
    "vgg16_freeze_best.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fine-Tuning the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 47ms/step - accuracy: 0.9532 - loss: 0.7321 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 2/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.9624 - loss: 0.6000 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 3/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 43ms/step - accuracy: 0.9551 - loss: 0.7007 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 4/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9579 - loss: 0.6667 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 5/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9662 - loss: 0.5434 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 6/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9467 - loss: 0.8380 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 7/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 37ms/step - accuracy: 0.9527 - loss: 0.7093 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 8/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 37ms/step - accuracy: 0.9562 - loss: 0.6778 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 9/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.9655 - loss: 0.5000 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 10/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 40ms/step - accuracy: 0.9552 - loss: 0.6766 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 11/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9608 - loss: 0.6105 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 12/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9633 - loss: 0.5878 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 13/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9597 - loss: 0.6437 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 14/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9626 - loss: 0.5910 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 15/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.9576 - loss: 0.6504 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 16/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9498 - loss: 0.7977 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 17/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9570 - loss: 0.6652 - val_accuracy: 0.9800 - val_loss: 0.3213\n",
      "Epoch 18/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 45ms/step - accuracy: 0.9626 - loss: 0.5986 - val_accuracy: 0.9800 - val_loss: 0.3161\n",
      "Epoch 19/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 44ms/step - accuracy: 0.9563 - loss: 0.6827 - val_accuracy: 0.9800 - val_loss: 0.3147\n",
      "Epoch 20/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 45ms/step - accuracy: 0.9615 - loss: 0.6141 - val_accuracy: 0.9800 - val_loss: 0.3140\n",
      "Epoch 21/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9602 - loss: 0.6330 - val_accuracy: 0.9800 - val_loss: 0.3148\n",
      "Epoch 22/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9635 - loss: 0.5584 - val_accuracy: 0.9800 - val_loss: 0.3072\n",
      "Epoch 23/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 43ms/step - accuracy: 0.9548 - loss: 0.7250 - val_accuracy: 0.9810 - val_loss: 0.3054\n",
      "Epoch 24/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 40ms/step - accuracy: 0.9620 - loss: 0.5819 - val_accuracy: 0.9810 - val_loss: 0.3057\n",
      "Epoch 25/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9580 - loss: 0.6321 - val_accuracy: 0.9800 - val_loss: 0.3063\n",
      "Epoch 26/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9606 - loss: 0.6211 - val_accuracy: 0.9810 - val_loss: 0.3053\n",
      "Epoch 27/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9619 - loss: 0.5875 - val_accuracy: 0.9810 - val_loss: 0.3054\n",
      "Epoch 28/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.9546 - loss: 0.7095 - val_accuracy: 0.9810 - val_loss: 0.3054\n",
      "Epoch 29/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9566 - loss: 0.6980 - val_accuracy: 0.9810 - val_loss: 0.3053\n",
      "Epoch 30/30\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 42ms/step - accuracy: 0.9521 - loss: 0.7495 - val_accuracy: 0.9810 - val_loss: 0.3053\n"
     ]
    }
   ],
   "source": [
    "# Unfreezing later layers for fine-tuning\n",
    "vgg16_base.trainable = True\n",
    "for layer in vgg16_base.layers[:-4]:\n",
    "    layer.trainable = False\n",
    "\n",
    "vgg16_freeze_model.compile(optimizer=keras.optimizers.RMSprop(learning_rate=1e-5),\n",
    "                            loss='binary_crossentropy',\n",
    "                            metrics=['accuracy']\n",
    ")\n",
    "\n",
    "vgg16_finetune_history = vgg16_freeze_model.fit(\n",
    "                        train_dataset,\n",
    "                        epochs = 30,\n",
    "                        validation_data = validation_dataset,\n",
    "                        callbacks = checkpoint_save(\"vgg16_finetune.keras\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Viualizing Model-training History"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 25ms/step - accuracy: 0.9694 - loss: 0.4538\n",
      "Test accuracy: 0.974\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"vgg16_freeze\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"vgg16_freeze\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)        </span>┃<span style=\"font-weight: bold\"> Output Shape      </span>┃<span style=\"font-weight: bold\">    Param # </span>┃<span style=\"font-weight: bold\"> Connected to      </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_6       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_11 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cast</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ input_layer_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ cast_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_1          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_2          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ sequential[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GetItem</span>)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ stack_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Stack</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ get_item[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],   │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │ get_item_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>], │\n",
       "│                     │                   │            │ get_item_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">180</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ stack_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ vgg16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Functional</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> │ add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]         │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ flatten_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12800</span>)     │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ vgg16[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │  <span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,056</span> │ flatten_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ dense_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)         │        <span style=\"color: #00af00; text-decoration-color: #00af00\">257</span> │ dropout_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)       \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m   Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to     \u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_6       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
       "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ cast_11 (\u001b[38;5;33mCast\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ input_layer_6[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ sequential          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ cast_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n",
       "│ (\u001b[38;5;33mSequential\u001b[0m)        │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item (\u001b[38;5;33mGetItem\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_1          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mGetItem\u001b[0m)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ get_item_2          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ sequential[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mGetItem\u001b[0m)           │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ stack_2 (\u001b[38;5;33mStack\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ get_item[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],   │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │ get_item_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
       "│                     │                   │            │ get_item_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (\u001b[38;5;33mAdd\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m180\u001b[0m, \u001b[38;5;34m180\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ stack_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n",
       "│                     │ \u001b[38;5;34m3\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ vgg16 (\u001b[38;5;33mFunctional\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m14,714,688\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]         │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ flatten_3 (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12800\u001b[0m)     │          \u001b[38;5;34m0\u001b[0m │ vgg16[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │  \u001b[38;5;34m3,277,056\u001b[0m │ flatten_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ dense_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense_5 (\u001b[38;5;33mDense\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)         │        \u001b[38;5;34m257\u001b[0m │ dropout_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">21,269,320</span> (81.14 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m21,269,320\u001b[0m (81.14 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,313</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m3,277,313\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,714,688</span> (56.13 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m14,714,688\u001b[0m (56.13 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">3,277,319</span> (12.50 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m3,277,319\u001b[0m (12.50 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the training history\n",
    "plot_history(vgg16_finetune_history)\n",
    "\n",
    "vgg16_finetune_best = keras.models.load_model(\"vgg16_finetune.keras\")\n",
    "vgg16_finetune_loss, vgg16_finetune_acc = vgg16_finetune_best.evaluate(test_dataset)\n",
    "print(f\"Test accuracy: {vgg16_finetune_acc:.3f}\")\n",
    "\n",
    "vgg16_finetune_best.summary()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "keras",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}