DL_Lab4.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyPcqQ5JMIe5Dv1M+c6iYUNe",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/megha-66/Deep-Learning-Lab-works-/blob/main/DL_Lab4.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "qHq_n-6G469M"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#LOAD the MNIST dataset\n",
        "mnist = tf.keras.datasets.mnist\n",
        "(train_images, train_labels), (test_images,test_labels) = mnist.load_data()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CjX3vy0X5lyJ",
        "outputId": "5c253005-67d0-461e-d5a4-fa69f068a804"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "\u001b[1m11490434/11490434\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Inspecting the images\n",
        "print(\"Shape of training images :\", train_images.shape)\n",
        "print(\"Shape of testing images : \", test_images.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PhkHtqfd6BH-",
        "outputId": "f58544b1-f0dc-4290-eacf-aa7bf4355a7e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape of training images : (60000, 28, 28)\n",
            "Shape of testing images :  (10000, 28, 28)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Visualizing a sample\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "sample = train_images[7] # any random image from training set\n",
        "plt.imshow(sample, cmap = 'gray')\n",
        "plt.title(f\"Label : {train_labels[7]}\")\n",
        "plt.show()\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "t3uPLd006g4r",
        "outputId": "075eb7d9-29f4-4cc3-98f7-f75c49646120"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "#Function to add Gaussian noise\n",
        "def add_gaussian_noise(images, mean = 0, stddev = 0.1):\n",
        "    noise  = np.random.normal(mean,stddev, images.shape)\n",
        "    noisy_images = noise + images\n",
        "    #Clip values to ensure pixel values are [0,1] range\n",
        "    noisy_images = np.clip(noisy_images,0.,1.)\n",
        "    return noisy_images\n",
        "\n",
        "#Adding gaussian noise to the training & testing datasets\n",
        "noisy_train_images = add_gaussian_noise(train_images, mean = 0 , stddev = 0.2)\n",
        "noisy_test_images = add_gaussian_noise(test_images, mean = 0, stddev = 0.2)\n",
        "\n",
        "#viewing sample noisy image\n",
        "plt.imshow(noisy_train_images[7], cmap='gray')\n",
        "plt.title(f\"Noisy Image - Label: {train_labels[7]}\")\n",
        "plt.show()\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "Y5SLsmN27EW2",
        "outputId": "5862f7a7-0515-4117-d2bb-69072b866c57"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras import layers, models\n"
      ],
      "metadata": {
        "id": "b8aoW0XY8T34"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def build_denoising_encoder():\n",
        "  #Encoder\n",
        "  input_img = layers.Input(shape = (28,28,1))\n",
        "  x = layers.Conv2D(32,(3,3), activation = 'relu', padding = 'same')(input_img)\n",
        "  x = layers.MaxPooling2D((2,2),padding = 'same')(x)\n",
        "  x = layers.Conv2D(64,(3,3), activation = 'relu', padding = 'same')(x)\n",
        "  encoded = layers.MaxPooling2D((2,2), padding = 'same')(x)\n",
        "\n",
        "  #Decoder\n",
        "  x = layers.Conv2DTranspose(64,(3,3), strides = 2, activation = 'relu', padding = 'same')(encoded)\n",
        "  x = layers.Conv2DTranspose(32,(3,3), strides = 2, activation = 'relu', padding = 'same')(x)\n",
        "  decoded = layers.Conv2D(1,(3,3), activation = 'sigmoid', padding = 'same')(x)\n",
        "\n",
        "  #Autoencoder label\n",
        "  autoencoder = models.Model(input_img, decoded)\n",
        "  autoencoder.compile(optimizer = 'adam', loss = 'mse')\n",
        "\n",
        "  return autoencoder\n",
        "\n",
        "#Build and summarize the model\n",
        "autoencoder = build_denoising_encoder()\n",
        "autoencoder.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 385
        },
        "id": "YNA0_y13-8jq",
        "outputId": "d14a680f-6eee-4f20-fd0d-e27d70481f02"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional\"\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: \"functional\"</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_2 (\u001b[38;5;33mInputLayer\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m1\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;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │             \u001b[38;5;34m320\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m32\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;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │          \u001b[38;5;34m18,496\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m64\u001b[0m)            │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_transpose (\u001b[38;5;33mConv2DTranspose\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │          \u001b[38;5;34m36,928\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_transpose_1 (\u001b[38;5;33mConv2DTranspose\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │          \u001b[38;5;34m18,464\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m1\u001b[0m)           │             \u001b[38;5;34m289\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_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</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\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │             <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</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\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)            │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_transpose (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2DTranspose</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_transpose_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2DTranspose</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,464</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\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">289</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m74,497\u001b[0m (291.00 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\">74,497</span> (291.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m74,497\u001b[0m (291.00 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\">74,497</span> (291.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\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\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "50cBykGMB9nj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder = build_denoising_encoder()\n",
        "early_stopping = EarlyStopping(monitor = 'val_loss', patience = 5, restore_best_weights = True)\n"
      ],
      "metadata": {
        "id": "O2HDeRZ9DJiu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Training the autoencoder\n",
        "history = autoencoder.fit(\n",
        "    noisy_train_images, train_images,\n",
        "    epochs = 2,\n",
        "    batch_size = 64,\n",
        "    shuffle = True,\n",
        "    validation_data = (noisy_test_images, test_images),\n",
        "    callbacks = [early_stopping]\n",
        "\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kcrbut_8Du7T",
        "outputId": "c855855d-0d1d-4c29-fe60-93048c1c27fb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/2\n",
            "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m169s\u001b[0m 180ms/step - loss: 7214.4971 - val_loss: 7342.7886\n",
            "Epoch 2/2\n",
            "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m166s\u001b[0m 177ms/step - loss: 7209.7959 - val_loss: 7342.7759\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#evaluating this trained model now on the test set\n",
        "test_loss = autoencoder.evaluate(noisy_test_images, test_images)\n",
        "print(\"Test Loss : \", test_loss)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l_gXtUN8EQFq",
        "outputId": "7a7e143c-819c-4dfe-b053-d54b7bec6375"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 22ms/step - loss: 6866.2910\n",
            "Test Loss :  7342.7744140625\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "\n",
        "random_indices = np.random.randint(0, len(noisy_train_images), 3)\n",
        "\n",
        "# Getting noisy images, clean images, and denoised outputs\n",
        "noisy_samples = noisy_train_images[random_indices]\n",
        "clean_samples = train_images[random_indices]\n",
        "denoised_samples = autoencoder.predict(noisy_samples)\n",
        "\n",
        "# Plotting the images for comparison\n",
        "plt.figure(figsize=(9, 3 * 3))\n",
        "\n",
        "for i in range(3):\n",
        "    # Noisy image\n",
        "    plt.subplot(3, 3, i * 3 + 1)\n",
        "    plt.imshow(noisy_samples[i].reshape(28, 28), cmap='gray')\n",
        "    plt.title(\"Noisy Image\")\n",
        "    plt.axis('off')\n",
        "\n",
        "    # Denoised image\n",
        "    plt.subplot(3, 3, i * 3 + 2)\n",
        "    plt.imshow(denoised_samples[i].reshape(28, 28), cmap='gray')\n",
        "    plt.title(\"Denoised Image\")\n",
        "    plt.axis('off')\n",
        "\n",
        "     # Original clean image\n",
        "    plt.subplot(3, 3, i * 3 + 3)\n",
        "    plt.imshow(clean_samples[i].reshape(28, 28), cmap='gray')\n",
        "    plt.title(\"Original Image\")\n",
        "    plt.axis('off')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 924
        },
        "id": "YpyWu3DCHPpG",
        "outputId": "a876802e-6633-4165-dea0-3276a6d6fbb5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 22ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 900x900 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "#Function to define salt and pepper noise\n",
        "def add_salt_and_pepper_noise(images, salt_prob=0.1, pepper_prob=0.01):\n",
        "    noisy_images = np.copy(images)\n",
        "    total_pixels = images.size\n",
        "\n",
        "    # Add salt noise (white pixels)\n",
        "    num_salt = int(total_pixels * salt_prob)\n",
        "    salt_coords = [np.random.randint(0, i - 1, num_salt) for i in images.shape]\n",
        "    noisy_images[salt_coords] = 1  # Set salt noise to 1 (white)\n",
        "\n",
        "    # Add pepper noise (black pixels)\n",
        "    num_pepper = int(total_pixels * pepper_prob)\n",
        "    pepper_coords = [np.random.randint(0, i - 1, num_pepper) for i in images.shape]\n",
        "    noisy_images[pepper_coords] = 0  # Set pepper noise to 0 (black)\n",
        "\n",
        "    return noisy_images\n",
        "\n",
        "# Adding salt-and-pepper noise to the training and testing datasets\n",
        "noisy_train_images_sp = add_salt_and_pepper_noise(train_images, salt_prob=0.9, pepper_prob=0)\n",
        "noisy_test_images_sp = add_salt_and_pepper_noise(test_images, salt_prob=0.01, pepper_prob=0)\n",
        "\n",
        "# View a sample noisy image with salt-and-pepper noise\n",
        "import matplotlib.pyplot as plt\n",
        "plt.imshow(noisy_train_images_sp[54].reshape(28, 28), cmap='gray')\n",
        "plt.title(\"Salt & Pepper Noisy Image\")\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "2YzRI_MAJ3Uc",
        "outputId": "e1395205-0741-42c7-da71-9b77bc55c09b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "def add_patchwise_noise(images, patch_size=(5, 5), num_patches=10, noise_std=0.1):\n",
        "    noisy_images = np.copy(images).astype(np.float32)\n",
        "    height, width = images.shape[1], images.shape[2]\n",
        "\n",
        "    for image in noisy_images:\n",
        "        for _ in range(num_patches):\n",
        "           # Randomly select a patch location\n",
        "             y = np.random.randint(0, height - patch_size[0])\n",
        "             x = np.random.randint(0, width - patch_size[1])\n",
        "\n",
        "            # Define the patch region\n",
        "             patch = image[y:y + patch_size[0], x:x + patch_size[1]]\n",
        "\n",
        "            # Add Gaussian noise to the patch\n",
        "             noise = np.random.normal(0, noise_std, patch.shape).astype(np.float32)\n",
        "             patch += noise\n",
        "\n",
        "             # Ensure pixel values are within [0, 1]\n",
        "             patch = np.clip(patch, 0., 1.)\n",
        "             image[y:y + patch_size[0], x:x + patch_size[1]] = patch\n",
        "\n",
        "    return noisy_images\n",
        "\n",
        "# Adding patchwise noise to the training and testing datasets\n",
        "noisy_train_images_patchwise = add_patchwise_noise(train_images, patch_size=(5, 5), num_patches=10, noise_std=0.1)\n",
        "noisy_test_images_patchwise = add_patchwise_noise(test_images, patch_size=(5, 5), num_patches=10, noise_std=0.1)\n",
        "\n",
        "# View a sample noisy image with patchwise noise\n",
        "import matplotlib.pyplot as plt\n",
        "plt.imshow(noisy_train_images_patchwise[0].reshape(28, 28), cmap='gray')\n",
        "plt.title(\"Patchwise Noisy Image\")\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "-lBISIy2LYkG",
        "outputId": "b8bf3804-a471-4ce2-bcd7-5713775e83c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}