AutoEncoder_TF_Keras_Fashion_Mnist.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "X2j5lCvoWr4d"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras import layers, losses\n",
        "from keras.datasets import fashion_mnist\n",
        "from keras.models import Model\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "y_train would be same as x_train because we want to reconstruct the original image"
      ],
      "metadata": {
        "id": "dqRRc1YMd5eF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "(x_train, _), (x_test, _) = fashion_mnist.load_data()\n",
        "x_train, x_test = x_train.astype('float32')/255., x_test.astype('float32')/255. #normalizing the images\n"
      ],
      "metadata": {
        "id": "Hnm1-ZMhXCl7"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(1,1)) #sample image\n",
        "plt.imshow(x_train[1])\n",
        "plt.gray()\n",
        "plt.axis('off')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 114
        },
        "id": "gzvTTgjgXds4",
        "outputId": "761d116b-f97b-4db3-d370-7273dffa94ae"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 100x100 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This autoencoder is made up of Dense layers"
      ],
      "metadata": {
        "id": "GjTPOm1CeJQM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Autoencoder(Model):\n",
        "    def __init__(self, latent_dim, shape):\n",
        "        super(Autoencoder, self).__init__()\n",
        "        self.latent_dim = latent_dim\n",
        "        self.shape = shape\n",
        "        self.encoder = tf.keras.Sequential([\n",
        "            layers.Flatten(),\n",
        "            layers.Dense(latent_dim, activation='relu')\n",
        "        ])\n",
        "        self.decoder = tf.keras.Sequential([\n",
        "            layers.Dense(tf.math.reduce_prod(shape), activation='sigmoid'),\n",
        "            layers.Reshape(shape)\n",
        "        ])\n",
        "\n",
        "    def call(self, x):\n",
        "        encoded = self.encoder(x)\n",
        "        decoded = self.decoder(encoded)\n",
        "        return decoded"
      ],
      "metadata": {
        "id": "_ci9GB9qY4XT"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "shape = x_train[0].shape\n",
        "latent_dim = 64\n",
        "print(shape)\n",
        "autoencoder = Autoencoder(latent_dim, shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mMIo-K_1a8qT",
        "outputId": "6bd69130-4ada-49ae-8bcd-7b0853843256"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(28, 28)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.compile(optimizer='adam', loss=losses.MeanSquaredError())\n",
        "autoencoder.fit(x_train, x_train, epochs=10, shuffle=True, validation_data=(x_test,x_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "voJ5aFV1bR2C",
        "outputId": "0481ca3c-867e-40f8-d265-aa1196776398"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1875/1875 [==============================] - 12s 6ms/step - loss: 0.0242 - val_loss: 0.0134\n",
            "Epoch 2/10\n",
            "1875/1875 [==============================] - 11s 6ms/step - loss: 0.0117 - val_loss: 0.0109\n",
            "Epoch 3/10\n",
            "1875/1875 [==============================] - 11s 6ms/step - loss: 0.0101 - val_loss: 0.0097\n",
            "Epoch 4/10\n",
            "1875/1875 [==============================] - 11s 6ms/step - loss: 0.0094 - val_loss: 0.0093\n",
            "Epoch 5/10\n",
            "1875/1875 [==============================] - 11s 6ms/step - loss: 0.0091 - val_loss: 0.0091\n",
            "Epoch 6/10\n",
            "1875/1875 [==============================] - 12s 6ms/step - loss: 0.0090 - val_loss: 0.0090\n",
            "Epoch 7/10\n",
            "1875/1875 [==============================] - 12s 7ms/step - loss: 0.0089 - val_loss: 0.0091\n",
            "Epoch 8/10\n",
            "1875/1875 [==============================] - 20s 11ms/step - loss: 0.0088 - val_loss: 0.0089\n",
            "Epoch 9/10\n",
            "1875/1875 [==============================] - 17s 9ms/step - loss: 0.0088 - val_loss: 0.0088\n",
            "Epoch 10/10\n",
            "1875/1875 [==============================] - 17s 9ms/step - loss: 0.0087 - val_loss: 0.0088\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7a590615b7c0>"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "encoded_imgs = autoencoder.encoder(x_test).numpy()\n",
        "decoded_imgs = autoencoder.decoder(encoded_imgs).numpy()"
      ],
      "metadata": {
        "id": "r-gShLavcFPZ"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(20,4))\n",
        "for i in range(10):\n",
        "    ax = plt.subplot(2,10,i+1)\n",
        "    plt.imshow(x_test[100*i])\n",
        "    plt.title('original')\n",
        "    plt.gray()\n",
        "    plt.axis('off')\n",
        "\n",
        "    ax = plt.subplot(2,10,10+i+1)\n",
        "    plt.imshow(decoded_imgs[100*i])\n",
        "    plt.title('reconstructed')\n",
        "    plt.gray()\n",
        "    plt.axis('off')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "V3YmcFxQcen8",
        "outputId": "e2af0bd6-e53d-40de-8724-3328971fc9f4"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 2000x400 with 20 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}