CS_E4875_course_(1)_(1).ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/LauriNieminen/LauriNieminen.github.io/blob/master/CS_E4875_course_(1)_(1).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b3d297ee-237f-412f-8ebf-8eca11320d63",
      "metadata": {
        "id": "b3d297ee-237f-412f-8ebf-8eca11320d63"
      },
      "source": [
        "## A Demo Notebook for Hyperspectral Super-Resolution Model (CS-E4875-course)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "f07299f0-1570-439c-b937-b8031b08f126",
      "metadata": {
        "id": "f07299f0-1570-439c-b937-b8031b08f126"
      },
      "source": [
        "### import Libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "a3376fb0-b90f-42e7-9f14-5bcc24d54942",
      "metadata": {
        "id": "a3376fb0-b90f-42e7-9f14-5bcc24d54942"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from scipy.io import loadmat\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras.layers import Input, Conv2D, UpSampling2D, BatchNormalization, Activation\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "from skimage.metrics import structural_similarity as ssim"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "e26e6f73-9c89-475c-a3f6-187483a413b1",
      "metadata": {
        "id": "e26e6f73-9c89-475c-a3f6-187483a413b1"
      },
      "source": [
        "### Preprocessing the dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "ca1d9099-b84c-4549-a6fe-08bb85048e1b",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ca1d9099-b84c-4549-a6fe-08bb85048e1b",
        "outputId": "0e790fd2-a515-410c-daa2-0effe40b54f4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Keys in loaded .mat file: dict_keys(['__header__', '__version__', '__globals__', 'paviaU'])\n",
            "Hyperspectral image shape: (610, 340, 103)\n",
            "X_train_hr shape: (44, 144, 144, 16)\n",
            "X_validation_hr shape: (6, 144, 144, 16)\n",
            "X_test_hr shape: (14, 144, 144, 16)\n",
            "X_train_lr shape: (44, 18, 18, 16)\n",
            "X_validation_lr shape: (6, 18, 18, 16)\n",
            "X_test_lr shape: (14, 18, 18, 16)\n",
            "Training Sample:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Validation Sample:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Sample:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "\n",
        "# Set random seeds for reproducibility\n",
        "np.random.seed(42)\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "# Load the Pavia dataset\n",
        "try:\n",
        "    data = loadmat(\"PaviaU.mat\")  # Ensure that the file path is correct\n",
        "except Exception as e:\n",
        "    raise ValueError(f\"Error loading .mat file: {e}\")\n",
        "\n",
        "# Access the hyperspectral image using the correct key 'paviaU'\n",
        "print(\"Keys in loaded .mat file:\", data.keys())\n",
        "if 'paviaU' in data:\n",
        "    hyperspectral_image = data['paviaU']\n",
        "else:\n",
        "    raise KeyError(\"'paviaU' not found in the .mat file.\")\n",
        "\n",
        "# Check the shape of the hyperspectral image\n",
        "print(\"Hyperspectral image shape:\", hyperspectral_image.shape)\n",
        "\n",
        "# Convert to float32 for TensorFlow operations\n",
        "hyperspectral_image = hyperspectral_image.astype(np.float32)\n",
        "\n",
        "# Parameters\n",
        "patch_size = (144, 144)  # Size of patches to extract\n",
        "test_size = 0.2  # Proportion of data for testing\n",
        "validation_size = 0.1  # Proportion of data for validation\n",
        "downscale_factor = 8  # Factor to downscale patches\n",
        "nodata_value = -1  # Value that indicates \"no data\"\n",
        "group_size = 16  # Group size for spectral bands\n",
        "overlap_size = 4  # Overlap size for grouped bands\n",
        "\n",
        "# Function to group bands into overlapping subgroups\n",
        "def group_bands_with_overlap(data, group_size=6, overlap_size=2):\n",
        "    height, width, bands = data.shape\n",
        "    step_size = group_size - overlap_size  # Calculate step size based on overlap\n",
        "    grouped_data = []\n",
        "\n",
        "    # Create overlapping groups of bands\n",
        "    for g in range(0, bands - group_size + 1, step_size):\n",
        "        group = data[:, :, g:g + group_size]\n",
        "        grouped_data.append(group)\n",
        "\n",
        "    return np.array(grouped_data)\n",
        "\n",
        "# Extract and downscale patches from hyperspectral data\n",
        "def extract_and_downscale_patches(data, patch_size, downscale_factor, nodata_value=0):\n",
        "    patches_hr = []\n",
        "    patches_lr = []\n",
        "    height, width, bands = data.shape\n",
        "\n",
        "    for i in range(0, height - patch_size[0] + 1, patch_size[0]):\n",
        "        for j in range(0, width - patch_size[1] + 1, patch_size[1]):\n",
        "            patch_hr = data[i:i + patch_size[0], j:j + patch_size[1], :]\n",
        "\n",
        "            # Check for nodata_value and skip patch extraction if present\n",
        "            if np.any(patch_hr == nodata_value):\n",
        "                continue\n",
        "\n",
        "            patch_lr = tf.image.resize(patch_hr,\n",
        "                                        [patch_size[0] // downscale_factor, patch_size[1] // downscale_factor],\n",
        "                                        method='bilinear')\n",
        "            patches_hr.append(patch_hr)\n",
        "            patches_lr.append(patch_lr.numpy())  # Convert tensor to numpy\n",
        "\n",
        "    return np.array(patches_hr), np.array(patches_lr)\n",
        "\n",
        "# Group bands into overlapping subgroups\n",
        "grouped_data = group_bands_with_overlap(hyperspectral_image, group_size=group_size, overlap_size=overlap_size)\n",
        "\n",
        "# Extract and downscale patches for all groups\n",
        "all_patches_hr = []\n",
        "all_patches_lr = []\n",
        "\n",
        "for group in grouped_data:\n",
        "    patches_hr, patches_lr = extract_and_downscale_patches(group, patch_size, downscale_factor, nodata_value=nodata_value)\n",
        "    all_patches_hr.append(patches_hr)\n",
        "    all_patches_lr.append(patches_lr)\n",
        "\n",
        "# Concatenate patches from all groups\n",
        "all_patches_hr = np.concatenate(all_patches_hr, axis=0)\n",
        "all_patches_lr = np.concatenate(all_patches_lr, axis=0)\n",
        "\n",
        "# Calculate the number of patches\n",
        "num_patches = len(all_patches_hr)\n",
        "\n",
        "# Calculate sizes for training, validation, and testing sets\n",
        "train_size = int((1 - test_size - validation_size) * num_patches)\n",
        "validation_size = int(validation_size * num_patches)\n",
        "test_size = num_patches - (train_size + validation_size)  # Explicit calculation of test size\n",
        "\n",
        "# Shuffle indices for splitting the data\n",
        "indices = np.arange(num_patches)\n",
        "np.random.shuffle(indices)\n",
        "all_patches_hr = all_patches_hr[indices]\n",
        "all_patches_lr = all_patches_lr[indices]\n",
        "\n",
        "# Split into training, validation, and testing sets\n",
        "X_train_hr, X_validation_hr, X_test_hr = np.split(all_patches_hr, [train_size, train_size + validation_size])\n",
        "X_train_lr, X_validation_lr, X_test_lr = np.split(all_patches_lr, [train_size, train_size + validation_size])\n",
        "\n",
        "# Print shapes to verify\n",
        "print(\"X_train_hr shape:\", X_train_hr.shape)\n",
        "print(\"X_validation_hr shape:\", X_validation_hr.shape)\n",
        "print(\"X_test_hr shape:\", X_test_hr.shape)\n",
        "\n",
        "print(\"X_train_lr shape:\", X_train_lr.shape)\n",
        "print(\"X_validation_lr shape:\", X_validation_lr.shape)\n",
        "print(\"X_test_lr shape:\", X_test_lr.shape)\n",
        "\n",
        "# Function to plot a high-resolution and low-resolution patch side by side\n",
        "def plot_sample(hr_patch, lr_patch, title):\n",
        "    plt.figure(figsize=(8, 4))\n",
        "\n",
        "    # Plot high-resolution patch\n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.imshow(hr_patch[:, :, 0], cmap='gray')  # Displaying the first spectral band\n",
        "    plt.title(f\"{title} High-Resolution Patch\")\n",
        "    plt.axis('off')\n",
        "\n",
        "    # Plot low-resolution patch\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.imshow(lr_patch[:, :, 0], cmap='gray')  # Displaying the first spectral band\n",
        "    plt.title(f\"{title} Low-Resolution Patch\")\n",
        "    plt.axis('off')\n",
        "\n",
        "    plt.show()\n",
        "\n",
        "# Display one training, validation, and test sample\n",
        "print(\"Training Sample:\")\n",
        "plot_sample(X_train_hr[0], X_train_lr[0], \"Training\")\n",
        "\n",
        "print(\"Validation Sample:\")\n",
        "plot_sample(X_validation_hr[0], X_validation_lr[0], \"Validation\")\n",
        "\n",
        "print(\"Test Sample:\")\n",
        "plot_sample(X_test_hr[0], X_test_lr[0], \"Test\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "3785fe77-1d80-4723-907a-94b8382bed4f",
      "metadata": {
        "id": "3785fe77-1d80-4723-907a-94b8382bed4f"
      },
      "source": [
        "## Insert the Model of your choice"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "17ec670d-00e0-41f3-85b1-0b949bff343f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "17ec670d-00e0-41f3-85b1-0b949bff343f",
        "outputId": "2f85b9a3-40ea-4c28-ed6a-f873bf9cf7bd"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"Simple_CNN_SR_Model\"\u001b[0m\n"
            ],
            "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: \"Simple_CNN_SR_Model\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "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;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m16\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;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │           \u001b[38;5;34m4,640\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │             \u001b[38;5;34m128\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation (\u001b[38;5;33mActivation\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\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;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │          \u001b[38;5;34m18,496\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_1                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │             \u001b[38;5;34m256\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_1 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\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;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │          \u001b[38;5;34m73,856\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_2                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │             \u001b[38;5;34m512\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_2 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m18\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d (\u001b[38;5;33mUpSampling2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m36\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;34m36\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │          \u001b[38;5;34m73,792\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_3                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │             \u001b[38;5;34m256\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_3 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d_1 (\u001b[38;5;33mUpSampling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m64\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;34m72\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │          \u001b[38;5;34m18,464\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_4                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │             \u001b[38;5;34m128\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_4 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d_2 (\u001b[38;5;33mUpSampling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m32\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;34m144\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │           \u001b[38;5;34m9,248\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_5                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m128\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_5 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m144\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;34m144\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m16\u001b[0m)        │           \u001b[38;5;34m4,624\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_6 (\u001b[38;5;33mActivation\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m144\u001b[0m, \u001b[38;5;34m16\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "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\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</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\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │           <span style=\"color: #00af00; text-decoration-color: #00af00\">4,640</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization                  │ (<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\">32</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</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\">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\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_1                │ (<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\">64</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</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\">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\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │          <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_2                │ (<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\">128</span>)         │             <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</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\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</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\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">73,792</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_3                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">36</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,464</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_4                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">72</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ up_sampling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</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\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │           <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ batch_normalization_5                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</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\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)        │           <span style=\"color: #00af00; text-decoration-color: #00af00\">4,624</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ activation_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m204,528\u001b[0m (798.94 KB)\n"
            ],
            "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\">204,528</span> (798.94 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m203,824\u001b[0m (796.19 KB)\n"
            ],
            "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\">203,824</span> (796.19 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m704\u001b[0m (2.75 KB)\n"
            ],
            "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\">704</span> (2.75 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "\n",
        "# Simple Convolutional Block\n",
        "def conv_block(x, filters, kernel_size=(3, 3)):\n",
        "    x = Conv2D(filters, kernel_size, padding='same')(x)\n",
        "    x = BatchNormalization()(x)\n",
        "    x = Activation('relu')(x)\n",
        "    return x\n",
        "\n",
        "# Upsampling Block\n",
        "def upsample_block(x, filters, scale=2):\n",
        "    x = UpSampling2D(size=(scale, scale), interpolation='bilinear')(x)\n",
        "    x = Conv2D(filters, (3, 3), padding='same')(x)\n",
        "    x = BatchNormalization()(x)\n",
        "    x = Activation('relu')(x)\n",
        "    return x\n",
        "\n",
        "# Build Simple CNN Model for Hyperspectral Super-Resolution\n",
        "def build_simple_cnn_sr_model(input_shape):\n",
        "    inputs = Input(shape=input_shape)\n",
        "\n",
        "    # Feature extraction layers\n",
        "    x = conv_block(inputs, filters=32)\n",
        "    x = conv_block(x, filters=64)\n",
        "    x = conv_block(x, filters=128)\n",
        "\n",
        "    # Upsampling layers\n",
        "    x = upsample_block(x, filters=64, scale=2)   # 2x upscaling\n",
        "    x = upsample_block(x, filters=32, scale=2)   # If you want upscale 4x upscaling then uncomment, if not then comment it\n",
        "    x = upsample_block(x, filters=32, scale=2)   # If you want upscale 8x upscaling, if not then comment it\n",
        "\n",
        "    # Output layer with the original number of channels\n",
        "    x_out = Conv2D(input_shape[-1], (3, 3), padding='same')(x)\n",
        "    x_out = Activation('linear')(x_out)\n",
        "\n",
        "    # Define the model\n",
        "    model = Model(inputs=inputs, outputs=x_out, name=\"Simple_CNN_SR_Model\")\n",
        "    return model\n",
        "\n",
        "# Define input shape for hyperspectral data\n",
        "input_shape = (18, 18, 16)  # Example input shape for low-resolution hyperspectral patch\n",
        "\n",
        "# Build the model\n",
        "simple_cnn_sr_model = build_simple_cnn_sr_model(input_shape)\n",
        "\n",
        "# Compile the model\n",
        "simple_cnn_sr_model.compile(optimizer='adam', loss='mse')\n",
        "\n",
        "# Display model summary\n",
        "simple_cnn_sr_model.summary()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6f33c583-b367-4511-82bd-31aeac3fd33c",
      "metadata": {
        "id": "6f33c583-b367-4511-82bd-31aeac3fd33c"
      },
      "source": [
        "## Train the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c8e7861c-aeff-4b13-a617-87763b3cd8aa",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c8e7861c-aeff-4b13-a617-87763b3cd8aa",
        "outputId": "be4f88ac-b70d-426f-b895-4fd834352d58"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 632ms/step - loss: 2760673.5000 - val_loss: 2651213.2500\n",
            "Epoch 2/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 449ms/step - loss: 2754430.7500 - val_loss: 2508082.5000\n",
            "Epoch 3/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 595ms/step - loss: 2747766.7500 - val_loss: 2468202.7500\n",
            "Epoch 4/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 471ms/step - loss: 2740048.0000 - val_loss: 2441586.0000\n",
            "Epoch 5/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 441ms/step - loss: 2730807.5000 - val_loss: 2478078.2500\n",
            "Epoch 6/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 589ms/step - loss: 2721374.7500 - val_loss: 2502083.7500\n",
            "Epoch 7/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 477ms/step - loss: 2711490.5000 - val_loss: 2502511.2500\n",
            "Epoch 8/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 443ms/step - loss: 2701193.5000 - val_loss: 2525451.7500\n",
            "Epoch 9/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 585ms/step - loss: 2690469.7500 - val_loss: 2536069.2500\n",
            "Epoch 10/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 484ms/step - loss: 2679303.5000 - val_loss: 2490661.7500\n",
            "Epoch 11/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 450ms/step - loss: 2667758.5000 - val_loss: 2491798.0000\n",
            "Epoch 12/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 579ms/step - loss: 2655722.7500 - val_loss: 2430585.2500\n",
            "Epoch 13/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 579ms/step - loss: 2643176.0000 - val_loss: 2471888.2500\n",
            "Epoch 14/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 454ms/step - loss: 2629967.5000 - val_loss: 2441918.7500\n",
            "Epoch 15/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 590ms/step - loss: 2616395.0000 - val_loss: 2370461.7500\n",
            "Epoch 16/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 443ms/step - loss: 2602216.7500 - val_loss: 2304077.2500\n",
            "Epoch 17/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 483ms/step - loss: 2587380.0000 - val_loss: 2367365.2500\n",
            "Epoch 18/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 460ms/step - loss: 2571871.7500 - val_loss: 2377442.5000\n",
            "Epoch 19/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 580ms/step - loss: 2555921.2500 - val_loss: 2376892.0000\n",
            "Epoch 20/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 479ms/step - loss: 2539686.0000 - val_loss: 2335541.0000\n",
            "Epoch 21/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 451ms/step - loss: 2522879.0000 - val_loss: 2340972.2500\n",
            "Epoch 22/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 585ms/step - loss: 2505506.5000 - val_loss: 2335056.0000\n",
            "Epoch 23/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 475ms/step - loss: 2487890.0000 - val_loss: 2251546.2500\n",
            "Epoch 24/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 437ms/step - loss: 2469711.7500 - val_loss: 2319252.7500\n",
            "Epoch 25/500\n",
            "\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 582ms/step - loss: 2451096.5000 - val_loss: 2284251.2500\n",
            "Epoch 26/500\n",
            "\u001b[1m 4/11\u001b[0m \u001b[32m━━━━━━━\u001b[0m\u001b[37m━━━━━━━━━━━━━\u001b[0m \u001b[1m2s\u001b[0m 421ms/step - loss: 2460813.0000"
          ]
        }
      ],
      "source": [
        "### # Define EarlyStopping callback\n",
        "early_stopping = EarlyStopping(monitor='val_loss', patience=30, restore_best_weights=True)\n",
        "\n",
        "# Train the model and get the training history with early stopping\n",
        "history = simple_cnn_sr_model.fit(\n",
        "    X_train_lr,\n",
        "    X_train_hr,\n",
        "    epochs=500,\n",
        "    batch_size=4,\n",
        "    validation_data=(X_validation_lr, X_validation_hr),\n",
        "    callbacks=[early_stopping]  # Add the early stopping callback here\n",
        ")\n",
        "\n",
        "# Visualize training and validation loss over epochs\n",
        "plt.figure(figsize=(10, 6))  # Adjust figure size if needed\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.grid(True)  # Add grid for better readability\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "277b5e00-cbbd-4e98-92a6-b05b6731e330",
      "metadata": {
        "id": "277b5e00-cbbd-4e98-92a6-b05b6731e330"
      },
      "source": [
        "## Report all the metrices such as PSNR, SSIM, SAM, ERGAS etc.,"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2c009d27-7a75-4cbe-b896-b4995c4576c5",
      "metadata": {
        "id": "2c009d27-7a75-4cbe-b896-b4995c4576c5",
        "outputId": "e3ae6389-fc75-4acb-b585-39a246681e96"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 1s 522ms/step\n",
            "Average PSNR on the test set: 30.221104962485178\n",
            "Average SSIM on the test set: 0.9100986\n",
            "Average SAM on the test set (in degrees): 4.692667778387005\n",
            "Average Correlation Coefficient on the test set: 0.945538293669639\n",
            "Average ERGAS on the test set: 9.15514410078758\n",
            "Average RMSE: 0.031690743\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Function to calculate PSNR without normalization\n",
        "def psnr(y_true, y_pred):\n",
        "    max_pixel = np.max(y_true)  # Use the actual max value from y_true\n",
        "    mse = np.mean((y_true - y_pred) ** 2)\n",
        "    if mse == 0:  # Avoid log of zero\n",
        "        return float('inf')\n",
        "    psnr_value = 20 * np.log10(max_pixel / np.sqrt(mse))\n",
        "    return psnr_value\n",
        "\n",
        "# Function to calculate SSIM with channel_axis\n",
        "def ssim_value(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(f\"Shape mismatch: y_true shape {y_true.shape} vs y_pred shape {y_pred.shape}\")\n",
        "\n",
        "    data_range = y_true.max() - y_true.min()  # Calculate data range from y_true\n",
        "    ssim_val = ssim(y_true, y_pred, data_range=data_range, channel_axis=-1)\n",
        "    return ssim_val\n",
        "\n",
        "# Function to calculate Correlation Coefficient\n",
        "def correlation_coefficient(y_true, y_pred):\n",
        "    y_true_flat = y_true.flatten()\n",
        "    y_pred_flat = y_pred.flatten()\n",
        "    corr_matrix = np.corrcoef(y_true_flat, y_pred_flat)\n",
        "    corr_value = corr_matrix[0, 1]\n",
        "    return corr_value\n",
        "\n",
        "# Function to calculate Spectral Angle Mapper (SAM) in degrees\n",
        "def sam(y_true, y_pred):\n",
        "    y_true_reshaped = y_true.reshape(-1, y_true.shape[-1])\n",
        "    y_pred_reshaped = y_pred.reshape(-1, y_pred.shape[-1])\n",
        "\n",
        "    non_zero_mask = (np.linalg.norm(y_true_reshaped, axis=1) > 1e-10) & (np.linalg.norm(y_pred_reshaped, axis=1) > 1e-10)\n",
        "    dot_product = np.sum(y_true_reshaped[non_zero_mask] * y_pred_reshaped[non_zero_mask], axis=1)\n",
        "    norm_true = np.linalg.norm(y_true_reshaped[non_zero_mask], axis=1)\n",
        "    norm_pred = np.linalg.norm(y_pred_reshaped[non_zero_mask], axis=1)\n",
        "\n",
        "    with np.errstate(divide='ignore', invalid='ignore'):\n",
        "        angles = np.arccos(np.clip(dot_product / (norm_true * norm_pred), -1.0, 1.0))\n",
        "\n",
        "    if angles.size > 0:\n",
        "        sam_value_degrees = np.mean(angles) * (180 / np.pi)\n",
        "    else:\n",
        "        sam_value_degrees = 0\n",
        "\n",
        "    return sam_value_degrees\n",
        "\n",
        "# Function to normalize the images\n",
        "def normalize(image):\n",
        "    min_val = np.min(image)\n",
        "    max_val = np.max(image)\n",
        "    return (image - min_val) / (max_val - min_val)  # Normalize to [0, 1]\n",
        "\n",
        "# Function to calculate Root Mean Squared Error (RMSE) for hyperspectral images (normalized)\n",
        "def rmse_bandwise(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(\"Shape mismatch between true and predicted images.\")\n",
        "\n",
        "    bands = y_true.shape[-1]\n",
        "    rmse_per_band = []\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)\n",
        "        rmse_band_value = np.sqrt(mse_band)\n",
        "        rmse_per_band.append(rmse_band_value)\n",
        "\n",
        "    # Normalize RMSE by the maximum value in y_true across all bands\n",
        "    max_value = np.max(y_true)\n",
        "    normalized_rmse = np.mean(rmse_per_band) / max_value\n",
        "    return normalized_rmse\n",
        "\n",
        "# Function to calculate ERGAS\n",
        "def ergas(y_true, y_pred, scale):\n",
        "    bands = y_true.shape[-1]\n",
        "    ergas_value = 0\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "        mean_band_true = np.mean(band_true)\n",
        "\n",
        "        # Calculate RMSE for the band without using a separate function\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)  # Mean Squared Error for the band\n",
        "        rmse_band = np.sqrt(mse_band)  # Root Mean Squared Error for the band\n",
        "\n",
        "        ergas_value += (rmse_band / mean_band_true) ** 2\n",
        "\n",
        "    ergas_value = 100 * (1 / scale) * np.sqrt(ergas_value / bands)\n",
        "    return ergas_value\n",
        "\n",
        "# Assuming hybrid_sr_model is trained, and X_test_lr, X_test_hr are defined\n",
        "predicted_hr_images = simple_cnn_sr_model.predict(X_test_lr)\n",
        "downscale_factor = 2 # ERGAS downscale factor\n",
        "\n",
        "# Validate shapes match for test and predictions\n",
        "if predicted_hr_images.shape != X_test_hr.shape:\n",
        "    raise ValueError(f\"Shape mismatch: predicted_hr_images shape {predicted_hr_images.shape} vs X_test_hr shape {X_test_hr.shape}\")\n",
        "\n",
        "# Calculate metrics per test sample\n",
        "psnr_values, ssim_values, cc_values, sam_values, ergas_values, rmse_values = [], [], [], [], [], []\n",
        "\n",
        "for i in range(len(X_test_hr)):\n",
        "    psnr_values.append(psnr(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ssim_values.append(ssim_value(X_test_hr[i], predicted_hr_images[i]))\n",
        "    cc_values.append(correlation_coefficient(X_test_hr[i], predicted_hr_images[i]))\n",
        "    sam_values.append(sam(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ergas_values.append(ergas(X_test_hr[i], predicted_hr_images[i], downscale_factor))\n",
        "    rmse_values.append(rmse_bandwise(X_test_hr[i], predicted_hr_images[i]))\n",
        "\n",
        "# Average metrics\n",
        "average_psnr = np.mean(psnr_values)\n",
        "average_ssim = np.mean(ssim_values)\n",
        "average_cc = np.mean(cc_values)\n",
        "average_sam = np.mean(sam_values)\n",
        "average_ergas = np.mean(ergas_values)\n",
        "average_rmse = np.mean(rmse_values)\n",
        "\n",
        "# Print the evaluation metrics\n",
        "print(\"Average PSNR on the test set:\", average_psnr)\n",
        "print(\"Average SSIM on the test set:\", average_ssim)\n",
        "print(\"Average SAM on the test set (in degrees):\", average_sam)\n",
        "print(\"Average Correlation Coefficient on the test set:\", average_cc)\n",
        "print(\"Average ERGAS on the test set:\", average_ergas)\n",
        "print(\"Average RMSE:\", average_rmse)  # Indicate RMSE is normalized\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b7f41667-b9dc-4c1b-a6f3-742450f4aec4",
      "metadata": {
        "id": "b7f41667-b9dc-4c1b-a6f3-742450f4aec4"
      },
      "source": [
        "## Scale size 4"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d8c41a19-a2ef-4ddb-b2f3-e0aceeb669d9",
      "metadata": {
        "id": "d8c41a19-a2ef-4ddb-b2f3-e0aceeb669d9",
        "outputId": "f28b208b-8cb7-46ae-90b5-a32a0ca35208"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 411ms/step\n",
            "Average PSNR on the test set: 27.6015841960907\n",
            "Average SSIM on the test set: 0.80812085\n",
            "Average SAM on the test set (in degrees): 4.696001886443281\n",
            "Average Correlation Coefficient on the test set: 0.9069213991165244\n",
            "Average ERGAS on the test set: 6.148700758030372\n",
            "Average RMSE: 0.04290982\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Function to calculate PSNR without normalization\n",
        "def psnr(y_true, y_pred):\n",
        "    max_pixel = np.max(y_true)  # Use the actual max value from y_true\n",
        "    mse = np.mean((y_true - y_pred) ** 2)\n",
        "    if mse == 0:  # Avoid log of zero\n",
        "        return float('inf')\n",
        "    psnr_value = 20 * np.log10(max_pixel / np.sqrt(mse))\n",
        "    return psnr_value\n",
        "\n",
        "# Function to calculate SSIM with channel_axis\n",
        "def ssim_value(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(f\"Shape mismatch: y_true shape {y_true.shape} vs y_pred shape {y_pred.shape}\")\n",
        "\n",
        "    data_range = y_true.max() - y_true.min()  # Calculate data range from y_true\n",
        "    ssim_val = ssim(y_true, y_pred, data_range=data_range, channel_axis=-1)\n",
        "    return ssim_val\n",
        "\n",
        "# Function to calculate Correlation Coefficient\n",
        "def correlation_coefficient(y_true, y_pred):\n",
        "    y_true_flat = y_true.flatten()\n",
        "    y_pred_flat = y_pred.flatten()\n",
        "    corr_matrix = np.corrcoef(y_true_flat, y_pred_flat)\n",
        "    corr_value = corr_matrix[0, 1]\n",
        "    return corr_value\n",
        "\n",
        "# Function to calculate Spectral Angle Mapper (SAM) in degrees\n",
        "def sam(y_true, y_pred):\n",
        "    y_true_reshaped = y_true.reshape(-1, y_true.shape[-1])\n",
        "    y_pred_reshaped = y_pred.reshape(-1, y_pred.shape[-1])\n",
        "\n",
        "    non_zero_mask = (np.linalg.norm(y_true_reshaped, axis=1) > 1e-10) & (np.linalg.norm(y_pred_reshaped, axis=1) > 1e-10)\n",
        "    dot_product = np.sum(y_true_reshaped[non_zero_mask] * y_pred_reshaped[non_zero_mask], axis=1)\n",
        "    norm_true = np.linalg.norm(y_true_reshaped[non_zero_mask], axis=1)\n",
        "    norm_pred = np.linalg.norm(y_pred_reshaped[non_zero_mask], axis=1)\n",
        "\n",
        "    with np.errstate(divide='ignore', invalid='ignore'):\n",
        "        angles = np.arccos(np.clip(dot_product / (norm_true * norm_pred), -1.0, 1.0))\n",
        "\n",
        "    if angles.size > 0:\n",
        "        sam_value_degrees = np.mean(angles) * (180 / np.pi)\n",
        "    else:\n",
        "        sam_value_degrees = 0\n",
        "\n",
        "    return sam_value_degrees\n",
        "\n",
        "# Function to normalize the images\n",
        "def normalize(image):\n",
        "    min_val = np.min(image)\n",
        "    max_val = np.max(image)\n",
        "    return (image - min_val) / (max_val - min_val)  # Normalize to [0, 1]\n",
        "\n",
        "# Function to calculate Root Mean Squared Error (RMSE) for hyperspectral images (normalized)\n",
        "def rmse_bandwise(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(\"Shape mismatch between true and predicted images.\")\n",
        "\n",
        "    bands = y_true.shape[-1]\n",
        "    rmse_per_band = []\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)\n",
        "        rmse_band_value = np.sqrt(mse_band)\n",
        "        rmse_per_band.append(rmse_band_value)\n",
        "\n",
        "    # Normalize RMSE by the maximum value in y_true across all bands\n",
        "    max_value = np.max(y_true)\n",
        "    normalized_rmse = np.mean(rmse_per_band) / max_value\n",
        "    return normalized_rmse\n",
        "\n",
        "# Function to calculate ERGAS\n",
        "def ergas(y_true, y_pred, scale):\n",
        "    bands = y_true.shape[-1]\n",
        "    ergas_value = 0\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "        mean_band_true = np.mean(band_true)\n",
        "\n",
        "        # Calculate RMSE for the band without using a separate function\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)  # Mean Squared Error for the band\n",
        "        rmse_band = np.sqrt(mse_band)  # Root Mean Squared Error for the band\n",
        "\n",
        "        ergas_value += (rmse_band / mean_band_true) ** 2\n",
        "\n",
        "    ergas_value = 100 * (1 / scale) * np.sqrt(ergas_value / bands)\n",
        "    return ergas_value\n",
        "\n",
        "# Assuming hybrid_sr_model is trained, and X_test_lr, X_test_hr are defined\n",
        "predicted_hr_images = simple_cnn_sr_model.predict(X_test_lr)\n",
        "downscale_factor = 4 # ERGAS downscale factor\n",
        "\n",
        "# Validate shapes match for test and predictions\n",
        "if predicted_hr_images.shape != X_test_hr.shape:\n",
        "    raise ValueError(f\"Shape mismatch: predicted_hr_images shape {predicted_hr_images.shape} vs X_test_hr shape {X_test_hr.shape}\")\n",
        "\n",
        "# Calculate metrics per test sample\n",
        "psnr_values, ssim_values, cc_values, sam_values, ergas_values, rmse_values = [], [], [], [], [], []\n",
        "\n",
        "for i in range(len(X_test_hr)):\n",
        "    psnr_values.append(psnr(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ssim_values.append(ssim_value(X_test_hr[i], predicted_hr_images[i]))\n",
        "    cc_values.append(correlation_coefficient(X_test_hr[i], predicted_hr_images[i]))\n",
        "    sam_values.append(sam(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ergas_values.append(ergas(X_test_hr[i], predicted_hr_images[i], downscale_factor))\n",
        "    rmse_values.append(rmse_bandwise(X_test_hr[i], predicted_hr_images[i]))\n",
        "\n",
        "# Average metrics\n",
        "average_psnr = np.mean(psnr_values)\n",
        "average_ssim = np.mean(ssim_values)\n",
        "average_cc = np.mean(cc_values)\n",
        "average_sam = np.mean(sam_values)\n",
        "average_ergas = np.mean(ergas_values)\n",
        "average_rmse = np.mean(rmse_values)\n",
        "\n",
        "# Print the evaluation metrics\n",
        "print(\"Average PSNR on the test set:\", average_psnr)\n",
        "print(\"Average SSIM on the test set:\", average_ssim)\n",
        "print(\"Average SAM on the test set (in degrees):\", average_sam)\n",
        "print(\"Average Correlation Coefficient on the test set:\", average_cc)\n",
        "print(\"Average ERGAS on the test set:\", average_ergas)\n",
        "print(\"Average RMSE:\", average_rmse)  # Indicate RMSE is normalized\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "ba86ee53-8f2c-4ecc-90dc-a196b7fd9d5d",
      "metadata": {
        "id": "ba86ee53-8f2c-4ecc-90dc-a196b7fd9d5d"
      },
      "source": [
        "## Scale size 8"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ce8e3bc9-0613-45ea-8843-2d805f2117ec",
      "metadata": {
        "id": "ce8e3bc9-0613-45ea-8843-2d805f2117ec",
        "outputId": "4c4af6b0-4747-4e10-8e2b-efd4c548d263"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 370ms/step\n",
            "Average PSNR on the test set: 26.855959381375992\n",
            "Average SSIM on the test set: 0.76512194\n",
            "Average SAM on the test set (in degrees): 4.698979681215868\n",
            "Average Correlation Coefficient on the test set: 0.8871284486519996\n",
            "Average ERGAS on the test set: 3.3437867006214357\n",
            "Average RMSE: 0.047208507\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Function to calculate PSNR without normalization\n",
        "def psnr(y_true, y_pred):\n",
        "    max_pixel = np.max(y_true)  # Use the actual max value from y_true\n",
        "    mse = np.mean((y_true - y_pred) ** 2)\n",
        "    if mse == 0:  # Avoid log of zero\n",
        "        return float('inf')\n",
        "    psnr_value = 20 * np.log10(max_pixel / np.sqrt(mse))\n",
        "    return psnr_value\n",
        "\n",
        "# Function to calculate SSIM with channel_axis\n",
        "def ssim_value(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(f\"Shape mismatch: y_true shape {y_true.shape} vs y_pred shape {y_pred.shape}\")\n",
        "\n",
        "    data_range = y_true.max() - y_true.min()  # Calculate data range from y_true\n",
        "    ssim_val = ssim(y_true, y_pred, data_range=data_range, channel_axis=-1)\n",
        "    return ssim_val\n",
        "\n",
        "# Function to calculate Correlation Coefficient\n",
        "def correlation_coefficient(y_true, y_pred):\n",
        "    y_true_flat = y_true.flatten()\n",
        "    y_pred_flat = y_pred.flatten()\n",
        "    corr_matrix = np.corrcoef(y_true_flat, y_pred_flat)\n",
        "    corr_value = corr_matrix[0, 1]\n",
        "    return corr_value\n",
        "\n",
        "# Function to calculate Spectral Angle Mapper (SAM) in degrees\n",
        "def sam(y_true, y_pred):\n",
        "    y_true_reshaped = y_true.reshape(-1, y_true.shape[-1])\n",
        "    y_pred_reshaped = y_pred.reshape(-1, y_pred.shape[-1])\n",
        "\n",
        "    non_zero_mask = (np.linalg.norm(y_true_reshaped, axis=1) > 1e-10) & (np.linalg.norm(y_pred_reshaped, axis=1) > 1e-10)\n",
        "    dot_product = np.sum(y_true_reshaped[non_zero_mask] * y_pred_reshaped[non_zero_mask], axis=1)\n",
        "    norm_true = np.linalg.norm(y_true_reshaped[non_zero_mask], axis=1)\n",
        "    norm_pred = np.linalg.norm(y_pred_reshaped[non_zero_mask], axis=1)\n",
        "\n",
        "    with np.errstate(divide='ignore', invalid='ignore'):\n",
        "        angles = np.arccos(np.clip(dot_product / (norm_true * norm_pred), -1.0, 1.0))\n",
        "\n",
        "    if angles.size > 0:\n",
        "        sam_value_degrees = np.mean(angles) * (180 / np.pi)\n",
        "    else:\n",
        "        sam_value_degrees = 0\n",
        "\n",
        "    return sam_value_degrees\n",
        "\n",
        "# Function to normalize the images\n",
        "def normalize(image):\n",
        "    min_val = np.min(image)\n",
        "    max_val = np.max(image)\n",
        "    return (image - min_val) / (max_val - min_val)  # Normalize to [0, 1]\n",
        "\n",
        "# Function to calculate Root Mean Squared Error (RMSE) for hyperspectral images (normalized)\n",
        "def rmse_bandwise(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(\"Shape mismatch between true and predicted images.\")\n",
        "\n",
        "    bands = y_true.shape[-1]\n",
        "    rmse_per_band = []\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)\n",
        "        rmse_band_value = np.sqrt(mse_band)\n",
        "        rmse_per_band.append(rmse_band_value)\n",
        "\n",
        "    # Normalize RMSE by the maximum value in y_true across all bands\n",
        "    max_value = np.max(y_true)\n",
        "    normalized_rmse = np.mean(rmse_per_band) / max_value\n",
        "    return normalized_rmse\n",
        "\n",
        "# Function to calculate ERGAS\n",
        "def ergas(y_true, y_pred, scale):\n",
        "    bands = y_true.shape[-1]\n",
        "    ergas_value = 0\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "        mean_band_true = np.mean(band_true)\n",
        "\n",
        "        # Calculate RMSE for the band without using a separate function\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)  # Mean Squared Error for the band\n",
        "        rmse_band = np.sqrt(mse_band)  # Root Mean Squared Error for the band\n",
        "\n",
        "        ergas_value += (rmse_band / mean_band_true) ** 2\n",
        "\n",
        "    ergas_value = 100 * (1 / scale) * np.sqrt(ergas_value / bands)\n",
        "    return ergas_value\n",
        "\n",
        "# Assuming hybrid_sr_model is trained, and X_test_lr, X_test_hr are defined\n",
        "predicted_hr_images = simple_cnn_sr_model.predict(X_test_lr)\n",
        "downscale_factor = 8 # ERGAS downscale factor\n",
        "\n",
        "# Validate shapes match for test and predictions\n",
        "if predicted_hr_images.shape != X_test_hr.shape:\n",
        "    raise ValueError(f\"Shape mismatch: predicted_hr_images shape {predicted_hr_images.shape} vs X_test_hr shape {X_test_hr.shape}\")\n",
        "\n",
        "# Calculate metrics per test sample\n",
        "psnr_values, ssim_values, cc_values, sam_values, ergas_values, rmse_values = [], [], [], [], [], []\n",
        "\n",
        "for i in range(len(X_test_hr)):\n",
        "    psnr_values.append(psnr(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ssim_values.append(ssim_value(X_test_hr[i], predicted_hr_images[i]))\n",
        "    cc_values.append(correlation_coefficient(X_test_hr[i], predicted_hr_images[i]))\n",
        "    sam_values.append(sam(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ergas_values.append(ergas(X_test_hr[i], predicted_hr_images[i], downscale_factor))\n",
        "    rmse_values.append(rmse_bandwise(X_test_hr[i], predicted_hr_images[i]))\n",
        "\n",
        "# Average metrics\n",
        "average_psnr = np.mean(psnr_values)\n",
        "average_ssim = np.mean(ssim_values)\n",
        "average_cc = np.mean(cc_values)\n",
        "average_sam = np.mean(sam_values)\n",
        "average_ergas = np.mean(ergas_values)\n",
        "average_rmse = np.mean(rmse_values)\n",
        "\n",
        "# Print the evaluation metrics\n",
        "print(\"Average PSNR on the test set:\", average_psnr)\n",
        "print(\"Average SSIM on the test set:\", average_ssim)\n",
        "print(\"Average SAM on the test set (in degrees):\", average_sam)\n",
        "print(\"Average Correlation Coefficient on the test set:\", average_cc)\n",
        "print(\"Average ERGAS on the test set:\", average_ergas)\n",
        "print(\"Average RMSE:\", average_rmse)  # Indicate RMSE is normalized\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "39b783ee-60ab-43d1-aa00-93c6e18a930b",
      "metadata": {
        "id": "39b783ee-60ab-43d1-aa00-93c6e18a930b",
        "outputId": "e1bb2c70-e891-4e82-f504-db68675f0163"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 450ms/step\n",
            "Average PSNR on the test set: 23.274510077067784\n",
            "Average SSIM on the test set: 0.53754276\n",
            "Average SAM on the test set (in degrees): 3.639469622860131\n",
            "Average Correlation Coefficient on the test set: 0.7245856279859101\n",
            "Average ERGAS on the test set: 5.066461490640917\n",
            "Average RMSE: 0.0710594\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Function to calculate PSNR without normalization\n",
        "def psnr(y_true, y_pred):\n",
        "    max_pixel = np.max(y_true)  # Use the actual max value from y_true\n",
        "    mse = np.mean((y_true - y_pred) ** 2)\n",
        "    if mse == 0:  # Avoid log of zero\n",
        "        return float('inf')\n",
        "    psnr_value = 20 * np.log10(max_pixel / np.sqrt(mse))\n",
        "    return psnr_value\n",
        "\n",
        "# Function to calculate SSIM with channel_axis\n",
        "def ssim_value(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(f\"Shape mismatch: y_true shape {y_true.shape} vs y_pred shape {y_pred.shape}\")\n",
        "\n",
        "    data_range = y_true.max() - y_true.min()  # Calculate data range from y_true\n",
        "    ssim_val = ssim(y_true, y_pred, data_range=data_range, channel_axis=-1)\n",
        "    return ssim_val\n",
        "\n",
        "# Function to calculate Correlation Coefficient\n",
        "def correlation_coefficient(y_true, y_pred):\n",
        "    y_true_flat = y_true.flatten()\n",
        "    y_pred_flat = y_pred.flatten()\n",
        "    corr_matrix = np.corrcoef(y_true_flat, y_pred_flat)\n",
        "    corr_value = corr_matrix[0, 1]\n",
        "    return corr_value\n",
        "\n",
        "# Function to calculate Spectral Angle Mapper (SAM) in degrees\n",
        "def sam(y_true, y_pred):\n",
        "    y_true_reshaped = y_true.reshape(-1, y_true.shape[-1])\n",
        "    y_pred_reshaped = y_pred.reshape(-1, y_pred.shape[-1])\n",
        "\n",
        "    non_zero_mask = (np.linalg.norm(y_true_reshaped, axis=1) > 1e-10) & (np.linalg.norm(y_pred_reshaped, axis=1) > 1e-10)\n",
        "    dot_product = np.sum(y_true_reshaped[non_zero_mask] * y_pred_reshaped[non_zero_mask], axis=1)\n",
        "    norm_true = np.linalg.norm(y_true_reshaped[non_zero_mask], axis=1)\n",
        "    norm_pred = np.linalg.norm(y_pred_reshaped[non_zero_mask], axis=1)\n",
        "\n",
        "    with np.errstate(divide='ignore', invalid='ignore'):\n",
        "        angles = np.arccos(np.clip(dot_product / (norm_true * norm_pred), -1.0, 1.0))\n",
        "\n",
        "    if angles.size > 0:\n",
        "        sam_value_degrees = np.mean(angles) * (180 / np.pi)\n",
        "    else:\n",
        "        sam_value_degrees = 0\n",
        "\n",
        "    return sam_value_degrees\n",
        "\n",
        "# Function to normalize the images\n",
        "def normalize(image):\n",
        "    min_val = np.min(image)\n",
        "    max_val = np.max(image)\n",
        "    return (image - min_val) / (max_val - min_val)  # Normalize to [0, 1]\n",
        "\n",
        "# Function to calculate Root Mean Squared Error (RMSE) for hyperspectral images (normalized)\n",
        "def rmse_bandwise(y_true, y_pred):\n",
        "    if y_true.shape != y_pred.shape:\n",
        "        raise ValueError(\"Shape mismatch between true and predicted images.\")\n",
        "\n",
        "    bands = y_true.shape[-1]\n",
        "    rmse_per_band = []\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)\n",
        "        rmse_band_value = np.sqrt(mse_band)\n",
        "        rmse_per_band.append(rmse_band_value)\n",
        "\n",
        "    # Normalize RMSE by the maximum value in y_true across all bands\n",
        "    max_value = np.max(y_true)\n",
        "    normalized_rmse = np.mean(rmse_per_band) / max_value\n",
        "    return normalized_rmse\n",
        "\n",
        "# Function to calculate ERGAS\n",
        "def ergas(y_true, y_pred, scale):\n",
        "    bands = y_true.shape[-1]\n",
        "    ergas_value = 0\n",
        "\n",
        "    for b in range(bands):\n",
        "        band_true = y_true[:, :, b]\n",
        "        band_pred = y_pred[:, :, b]\n",
        "        mean_band_true = np.mean(band_true)\n",
        "\n",
        "        # Calculate RMSE for the band without using a separate function\n",
        "        mse_band = np.mean((band_true - band_pred) ** 2)  # Mean Squared Error for the band\n",
        "        rmse_band = np.sqrt(mse_band)  # Root Mean Squared Error for the band\n",
        "\n",
        "        ergas_value += (rmse_band / mean_band_true) ** 2\n",
        "\n",
        "    ergas_value = 100 * (1 / scale) * np.sqrt(ergas_value / bands)\n",
        "    return ergas_value\n",
        "\n",
        "# Assuming hybrid_sr_model is trained, and X_test_lr, X_test_hr are defined\n",
        "predicted_hr_images = residual_cnn_sr_model.predict(X_test_lr)\n",
        "downscale_factor = 8 # ERGAS downscale factor\n",
        "\n",
        "# Validate shapes match for test and predictions\n",
        "if predicted_hr_images.shape != X_test_hr.shape:\n",
        "    raise ValueError(f\"Shape mismatch: predicted_hr_images shape {predicted_hr_images.shape} vs X_test_hr shape {X_test_hr.shape}\")\n",
        "\n",
        "# Calculate metrics per test sample\n",
        "psnr_values, ssim_values, cc_values, sam_values, ergas_values, rmse_values = [], [], [], [], [], []\n",
        "\n",
        "for i in range(len(X_test_hr)):\n",
        "    psnr_values.append(psnr(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ssim_values.append(ssim_value(X_test_hr[i], predicted_hr_images[i]))\n",
        "    cc_values.append(correlation_coefficient(X_test_hr[i], predicted_hr_images[i]))\n",
        "    sam_values.append(sam(X_test_hr[i], predicted_hr_images[i]))\n",
        "    ergas_values.append(ergas(X_test_hr[i], predicted_hr_images[i], downscale_factor))\n",
        "    rmse_values.append(rmse_bandwise(X_test_hr[i], predicted_hr_images[i]))\n",
        "\n",
        "# Average metrics\n",
        "average_psnr = np.mean(psnr_values)\n",
        "average_ssim = np.mean(ssim_values)\n",
        "average_cc = np.mean(cc_values)\n",
        "average_sam = np.mean(sam_values)\n",
        "average_ergas = np.mean(ergas_values)\n",
        "average_rmse = np.mean(rmse_values)\n",
        "\n",
        "# Print the evaluation metrics\n",
        "print(\"Average PSNR on the test set:\", average_psnr)\n",
        "print(\"Average SSIM on the test set:\", average_ssim)\n",
        "print(\"Average SAM on the test set (in degrees):\", average_sam)\n",
        "print(\"Average Correlation Coefficient on the test set:\", average_cc)\n",
        "print(\"Average ERGAS on the test set:\", average_ergas)\n",
        "print(\"Average RMSE:\", average_rmse)  # Indicate RMSE is normalized\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "1d10a855-e272-44b6-adfc-a25b94d3f709",
      "metadata": {
        "id": "1d10a855-e272-44b6-adfc-a25b94d3f709"
      },
      "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": "3.9.18"
    },
    "colab": {
      "provenance": [],
      "include_colab_link": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}