UNET_MODEL ANALYSIS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "88d1369a-b266-4571-bdc6-e23f256cfe64",
   "metadata": {},
   "source": [
    " UNET MODEL ANALYSIS"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b27e1d44-cda6-46d4-bcd0-5bf0d68e3c3a",
   "metadata": {},
   "source": [
    "IMPORTING ALL LIBRARIES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f9bc6bb5-890d-495e-8459-f6cddf933232",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import cv2\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.models import load_model\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, UpSampling2D, concatenate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d7e8e68-b0f4-4d16-a913-015e90f3ab13",
   "metadata": {},
   "source": [
    "LOADING THE PRETRAINED MODEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5e184e09-6b81-4c32-9b81-a77a2578edc4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    }
   ],
   "source": [
    "model = tf.keras.models.load_model('unet_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ad2fcf04-81b2-4fe6-9a8d-0fa22da3effb",
   "metadata": {},
   "outputs": [],
   "source": [
    "path_images = 'test/original_images'\n",
    "path_masks = 'test/binary_images'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b904db2-4cf6-4818-9cf9-0a06d026df7a",
   "metadata": {},
   "source": [
    "LAODING IAMGES "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "719c081f-901d-49af-b7b8-5690c79b78d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_images(image_folder, mask_folder):\n",
    "    images = []\n",
    "    masks = []\n",
    "\n",
    "    image_files = sorted([f for f in os.listdir(image_folder) if os.path.isfile(os.path.join(image_folder, f)) and f.endswith('.png')])\n",
    "    mask_files = sorted([f for f in os.listdir(mask_folder) if os.path.isfile(os.path.join(mask_folder, f)) and f.endswith('.png')])\n",
    "\n",
    "    if len(image_files) != len(mask_files):\n",
    "        print(f\"Number of images ({len(image_files)}) does not match number of masks ({len(mask_files)}).\")\n",
    "        return None, None\n",
    "\n",
    "    print(f\"Image Files: {image_files}\")\n",
    "    print(f\"Mask Files: {mask_files}\")\n",
    "\n",
    "    for img_file, mask_file in zip(image_files, mask_files):\n",
    "        img_path = os.path.join(image_folder, img_file)\n",
    "        mask_path = os.path.join(mask_folder, mask_file)\n",
    "\n",
    "        img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)\n",
    "        if img is None:\n",
    "            print(f\"Failed to read image {img_file}.\")\n",
    "            continue\n",
    "        \n",
    "        img = cv2.resize(img, (IMG_WIDTH, IMG_HEIGHT))\n",
    "        img = np.expand_dims(img, axis=-1) / 255.0  # Normalize and reshape\n",
    "        \n",
    "        mask = cv2.imread(mask_path, cv2.IMREAD_GRAYSCALE)\n",
    "        if mask is None:\n",
    "            print(f\"Failed to read mask {mask_file}.\")\n",
    "            continue\n",
    "        \n",
    "        mask = cv2.resize(mask, (IMG_WIDTH, IMG_HEIGHT))\n",
    "        mask = np.expand_dims(mask, axis=-1) / 255.0\n",
    "        \n",
    "        images.append(img)\n",
    "        masks.append(mask)\n",
    "\n",
    "    print(f\"Successfully loaded {len(images)} images and {len(masks)} masks.\")\n",
    "    return np.array(images), np.array(masks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cda39091-69c4-4f40-ae72-a4332bfc3243",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMG_HEIGHT, IMG_WIDTH = 128, 128\n",
    "IMG_CHANNELS = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c926fc49-4b86-4d2d-aeea-ee24a9cba3c6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image Files: ['0001.png', '0002.png', '0003.png', '0004.png', '0005.png', '0006.png', '0007.png', '0008.png', '0009.png', '0010.png', '0011.png', '0012.png', '0013.png', '0014.png', '0015.png', '0016.png', '0017.png', '0018.png', '0019.png', '0020.png', '0021.png', '0022.png', '0023.png', '0024.png', '0025.png', '0026.png', '0027.png', '0028.png', '0029.png', '0030.png', '0031.png', '0032.png', '0033.png', '0034.png', '0035.png', '0036.png', '0037.png', '0038.png', '0039.png', '0040.png', '0041.png', '0042.png', '0043.png', '0044.png', '0045.png', '0046.png', '0047.png', '0048.png', '0049.png', '0050.png', '0051.png', '0052.png', '0053.png', '0054.png', '0055.png', '0056.png', '0057.png', '0058.png', '0059.png', '0060.png', '0061.png', '0062.png', '0063.png', '0064.png', '0065.png', '0066.png', '0067.png', '0068.png', '0069.png', '0070.png', '0071.png', '0072.png', '0073.png', '0074.png', '0075.png', '0076.png', '0077.png', '0078.png', '0079.png', '0080.png', '0081.png', '0082.png', '0083.png', '0084.png', '0085.png', '0086.png', '0087.png', '0088.png', '0089.png', '0090.png', '0091.png', '0092.png', '0093.png', '0094.png', '0095.png', '0096.png', '0097.png', '0098.png', '0099.png', '0100.png', '0101.png', '0102.png', '0103.png', '0104.png', '0105.png', '0106.png', '0107.png', '0108.png', '0109.png', '0110.png', '0111.png', '0112.png', '0113.png', '0114.png', '0115.png', '0116.png', '0117.png', '0118.png', '0119.png', '0120.png', '0121.png', '0122.png', '0123.png', '0124.png', '0125.png', '0126.png', '0127.png', '0128.png', '0129.png', '0130.png', '0131.png', '0132.png', '0133.png', '0134.png', '0135.png', '0136.png', '0137.png', '0138.png', '0139.png', '0140.png', '0141.png', '0142.png', '0143.png', '0144.png', '0145.png', '0146.png', '0147.png', '0148.png', '0149.png', '0150.png', '0151.png', '0152.png', '0153.png', '0154.png', '0155.png', '0156.png', '0157.png', '0158.png', '0159.png', '0160.png', '0161.png', '0162.png', '0163.png', '0164.png', '0165.png', '0166.png', '0167.png', '0168.png', '0169.png', '0170.png', '0171.png', '0172.png', '0173.png', '0174.png', '0175.png', '0176.png', '0177.png', '0178.png', '0179.png', '0180.png', '0181.png', '0182.png', '0183.png', '0184.png', '0185.png', '0186.png', '0187.png', '0188.png', '0189.png', '0190.png', '0191.png', '0192.png', '0193.png', '0194.png', '0195.png', '0196.png', '0197.png', '0198.png', '0199.png', '0200.png', '0201.png', '0202.png', '0203.png', '0204.png', '0205.png', '0206.png', '0207.png', '0208.png', '0209.png', '0210.png', '0211.png', '0212.png', '0213.png', '0214.png', '0215.png', '0216.png', '0217.png', '0218.png', '0219.png', '0220.png', '0221.png', '0222.png', '0223.png', '0224.png', '0225.png', '0226.png', '0227.png', '0228.png', '0229.png', '0230.png', '0231.png', '0232.png', '0233.png', '0234.png', '0235.png', '0236.png', '0237.png', '0238.png', '0239.png', '0240.png', '0241.png', '0242.png', '0243.png', '0244.png', '0245.png', '0246.png', '0247.png', '0248.png', '0249.png', '0250.png']\n",
      "Mask Files: ['0001.png', '0002.png', '0003.png', '0004.png', '0005.png', '0006.png', '0007.png', '0008.png', '0009.png', '0010.png', '0011.png', '0012.png', '0013.png', '0014.png', '0015.png', '0016.png', '0017.png', '0018.png', '0019.png', '0020.png', '0021.png', '0022.png', '0023.png', '0024.png', '0025.png', '0026.png', '0027.png', '0028.png', '0029.png', '0030.png', '0031.png', '0032.png', '0033.png', '0034.png', '0035.png', '0036.png', '0037.png', '0038.png', '0039.png', '0040.png', '0041.png', '0042.png', '0043.png', '0044.png', '0045.png', '0046.png', '0047.png', '0048.png', '0049.png', '0050.png', '0051.png', '0052.png', '0053.png', '0054.png', '0055.png', '0056.png', '0057.png', '0058.png', '0059.png', '0060.png', '0061.png', '0062.png', '0063.png', '0064.png', '0065.png', '0066.png', '0067.png', '0068.png', '0069.png', '0070.png', '0071.png', '0072.png', '0073.png', '0074.png', '0075.png', '0076.png', '0077.png', '0078.png', '0079.png', '0080.png', '0081.png', '0082.png', '0083.png', '0084.png', '0085.png', '0086.png', '0087.png', '0088.png', '0089.png', '0090.png', '0091.png', '0092.png', '0093.png', '0094.png', '0095.png', '0096.png', '0097.png', '0098.png', '0099.png', '0100.png', '0101.png', '0102.png', '0103.png', '0104.png', '0105.png', '0106.png', '0107.png', '0108.png', '0109.png', '0110.png', '0111.png', '0112.png', '0113.png', '0114.png', '0115.png', '0116.png', '0117.png', '0118.png', '0119.png', '0120.png', '0121.png', '0122.png', '0123.png', '0124.png', '0125.png', '0126.png', '0127.png', '0128.png', '0129.png', '0130.png', '0131.png', '0132.png', '0133.png', '0134.png', '0135.png', '0136.png', '0137.png', '0138.png', '0139.png', '0140.png', '0141.png', '0142.png', '0143.png', '0144.png', '0145.png', '0146.png', '0147.png', '0148.png', '0149.png', '0150.png', '0151.png', '0152.png', '0153.png', '0154.png', '0155.png', '0156.png', '0157.png', '0158.png', '0159.png', '0160.png', '0161.png', '0162.png', '0163.png', '0164.png', '0165.png', '0166.png', '0167.png', '0168.png', '0169.png', '0170.png', '0171.png', '0172.png', '0173.png', '0174.png', '0175.png', '0176.png', '0177.png', '0178.png', '0179.png', '0180.png', '0181.png', '0182.png', '0183.png', '0184.png', '0185.png', '0186.png', '0187.png', '0188.png', '0189.png', '0190.png', '0191.png', '0192.png', '0193.png', '0194.png', '0195.png', '0196.png', '0197.png', '0198.png', '0199.png', '0200.png', '0201.png', '0202.png', '0203.png', '0204.png', '0205.png', '0206.png', '0207.png', '0208.png', '0209.png', '0210.png', '0211.png', '0212.png', '0213.png', '0214.png', '0215.png', '0216.png', '0217.png', '0218.png', '0219.png', '0220.png', '0221.png', '0222.png', '0223.png', '0224.png', '0225.png', '0226.png', '0227.png', '0228.png', '0229.png', '0230.png', '0231.png', '0232.png', '0233.png', '0234.png', '0235.png', '0236.png', '0237.png', '0238.png', '0239.png', '0240.png', '0241.png', '0242.png', '0243.png', '0244.png', '0245.png', '0246.png', '0247.png', '0248.png', '0249.png', '0250.png']\n",
      "Successfully loaded 250 images and 250 masks.\n"
     ]
    }
   ],
   "source": [
    "X_test,Y_test = load_images(path_images,path_masks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f86e6096-7cca-4560-8291-402ef94218f7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m8/8\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m34s\u001b[0m 4s/step\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c498672d-f51d-426e-9c51-5ce1b5402981",
   "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: \"functional_1\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"functional_1\"\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_1       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</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\">1</span>)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_19 (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">640</span> │ input_layer_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_20 (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │ conv2d_19[<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",
       "│ max_pooling2d_4     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_20[<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\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_21 (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │ max_pooling2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_22 (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ conv2d_21[<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",
       "│ max_pooling2d_5     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_22[<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\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_23 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │ max_pooling2d_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_24 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ conv2d_23[<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",
       "│ max_pooling2d_6     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_24[<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\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_25 (<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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,180,160</span> │ max_pooling2d_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_26 (<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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │ conv2d_25[<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\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_7     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_26[<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\">MaxPooling2D</span>)      │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_27 (<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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">4,719,616</span> │ max_pooling2d_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_28 (<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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">9,438,208</span> │ conv2d_27[<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\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_4     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_28[<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\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_4       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ up_sampling2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1536</span>)             │            │ conv2d_26[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_29 (<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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │  <span style=\"color: #00af00; text-decoration-color: #00af00\">7,078,400</span> │ concatenate_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_30 (<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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │ conv2d_29[<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\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_5     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_30[<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\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_5       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ up_sampling2d_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">768</span>)              │            │ conv2d_24[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_31 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,769,728</span> │ concatenate_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_32 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ conv2d_31[<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",
       "│ up_sampling2d_6     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_32[<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\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_6       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ up_sampling2d_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">384</span>)              │            │ conv2d_22[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_33 (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">442,496</span> │ concatenate_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_34 (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ conv2d_33[<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",
       "│ up_sampling2d_7     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_34[<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\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_7       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ up_sampling2d_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │ conv2d_20[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_35 (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │    <span style=\"color: #00af00; text-decoration-color: #00af00\">110,656</span> │ concatenate_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_36 (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │ conv2d_35[<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",
       "│ conv2d_37 (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>,  │         <span style=\"color: #00af00; text-decoration-color: #00af00\">65</span> │ conv2d_36[<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\">1</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_1       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
       "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;34m1\u001b[0m)                │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_19 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │        \u001b[38;5;34m640\u001b[0m │ input_layer_1[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_20 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │     \u001b[38;5;34m36,928\u001b[0m │ conv2d_19[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_4     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_20[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_21 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │     \u001b[38;5;34m73,856\u001b[0m │ max_pooling2d_4[\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_22 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │    \u001b[38;5;34m147,584\u001b[0m │ conv2d_21[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_5     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_22[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_23 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │    \u001b[38;5;34m295,168\u001b[0m │ max_pooling2d_5[\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_24 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │    \u001b[38;5;34m590,080\u001b[0m │ conv2d_23[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_6     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_24[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_25 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │  \u001b[38;5;34m1,180,160\u001b[0m │ max_pooling2d_6[\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_26 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │  \u001b[38;5;34m2,359,808\u001b[0m │ conv2d_25[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d_7     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ conv2d_26[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_27 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m,      │  \u001b[38;5;34m4,719,616\u001b[0m │ max_pooling2d_7[\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_28 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m,      │  \u001b[38;5;34m9,438,208\u001b[0m │ conv2d_27[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_4     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_28[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_4       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ up_sampling2d_4[\u001b[38;5;34m…\u001b[0m │\n",
       "│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m1536\u001b[0m)             │            │ conv2d_26[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_29 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │  \u001b[38;5;34m7,078,400\u001b[0m │ concatenate_4[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_30 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │  \u001b[38;5;34m2,359,808\u001b[0m │ conv2d_29[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_5     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_30[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_5       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ up_sampling2d_5[\u001b[38;5;34m…\u001b[0m │\n",
       "│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m768\u001b[0m)              │            │ conv2d_24[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_31 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │  \u001b[38;5;34m1,769,728\u001b[0m │ concatenate_5[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_32 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │    \u001b[38;5;34m590,080\u001b[0m │ conv2d_31[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_6     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ conv2d_32[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_6       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ up_sampling2d_6[\u001b[38;5;34m…\u001b[0m │\n",
       "│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m384\u001b[0m)              │            │ conv2d_22[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_33 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │    \u001b[38;5;34m442,496\u001b[0m │ concatenate_6[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_34 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m,    │    \u001b[38;5;34m147,584\u001b[0m │ conv2d_33[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ up_sampling2d_7     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ conv2d_34[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ concatenate_7       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ up_sampling2d_7[\u001b[38;5;34m…\u001b[0m │\n",
       "│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m192\u001b[0m)              │            │ conv2d_20[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_35 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │    \u001b[38;5;34m110,656\u001b[0m │ concatenate_7[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_36 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │     \u001b[38;5;34m36,928\u001b[0m │ conv2d_35[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_37 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m,  │         \u001b[38;5;34m65\u001b[0m │ conv2d_36[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│                     │ \u001b[38;5;34m1\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\">31,377,795</span> (119.70 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m31,377,795\u001b[0m (119.70 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\">31,377,793</span> (119.70 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m31,377,793\u001b[0m (119.70 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\">2</span> (12.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m2\u001b[0m (12.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d372f8f1-62d5-43e5-b9fc-5a155e97c3d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "492dfe22-7cbf-4860-9391-eeed5fd79ef1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m8/8\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 4s/step - accuracy: 0.9863 - loss: 0.0383\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.03861866518855095, 0.9859650731086731]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c86ea280-3f7a-476c-8c67-3494d1807ac4",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true_flat = Y_test.flatten()  \n",
    "y_pred_flat = predictions.flatten() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "96b6f238-9e33-42d7-8dc4-0c51169d30d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true_flat = (y_true_flat > 0.5).astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c8eb77c-db39-4cca-9d94-d08c77e89518",
   "metadata": {},
   "source": [
    "ROC CURVE "
   ]
  },
  {
   "cell_type": "raw",
   "id": "7cfd00cb-8585-44f3-9fd8-fa776e75fc1e",
   "metadata": {},
   "source": [
    "FORMULA:\n",
    "TPR = TP/TP+FN \n",
    "FPR = FP/FP+TN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "e6ebcb00-1028-4c26-b57b-e1f623e21cc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_true_flat, y_pred_flat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7013d466-b52b-4710-bb9e-2b2a9a5556ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "roc_auc = auc(fpr, tpr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3e1c94cc-c42d-4fb9-af38-11c5e472111f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(fpr, tpr, color='darkorange', lw=2, label=f'ROC curve (area = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC)')\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03fa3519-e8c4-47b9-897d-ccb62bd72de5",
   "metadata": {},
   "source": [
    "PRECISION - RECALL  F1 SCORE CURVE "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "4c0a5358-ecef-4b55-b80e-31c998dd75fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve, auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "4ba9e8f1-73c2-415f-a4fa-8825e16cff8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "precision, recall, thresholds = precision_recall_curve(y_true_flat, y_pred_flat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "df5de041-3092-4acb-9a9a-1d792907f86c",
   "metadata": {},
   "outputs": [],
   "source": [
    "pr_auc = auc(recall, precision)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "bcde10f3-3375-427f-8d35-0b12a0cdd1ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "f1_scores = 2 * (precision * recall) / (precision + recall + 1e-10)"
   ]
  },
  {
   "cell_type": "raw",
   "id": "931efc14-698d-4726-9206-3dfc706b9e65",
   "metadata": {},
   "source": [
    "PRECISION = TP/TP+FP\n",
    "RECALL = TP/TP+FN\n",
    "F1_SCORES = HARMONIC MEAN ( RECALL , PRECISION )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "93303487-681a-4010-89d3-fc4af50cac1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(thresholds, precision[:-1], label='Precision', marker='o')\n",
    "plt.plot(thresholds, recall[:-1], label='Recall', marker='s')\n",
    "plt.plot(thresholds, f1_scores[:-1], label='F1 Score', marker='^')\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('Score')\n",
    "plt.title('Precision, Recall, and F1 Score vs. Threshold')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4cef6c2-1a15-4712-aae7-9bbd8505c0e6",
   "metadata": {},
   "source": [
    "PIXEL ACCURACY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "34b08e85-c59a-471e-a1e3-796848bf40bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true_flat = Y_test.flatten()\n",
    "y_pred_flat = (predictions.flatten() > 0.5).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5b3c33be-8f0d-445d-a830-2455caf53608",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pixel Accuracy: 0.986\n"
     ]
    }
   ],
   "source": [
    "correct_pixels = np.sum(y_true_flat == y_pred_flat)\n",
    "total_pixels = y_true_flat.size\n",
    "\n",
    "pixel_accuracy = correct_pixels / total_pixels\n",
    "\n",
    "print(f\"Pixel Accuracy: {pixel_accuracy:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dbc2e22c-c73d-49da-b778-743cb9f58ee4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}