Convolutional AutoEncoder.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled16.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/haseeb1zahid/Deep-Learning-Projects/blob/master/Convolutional%20AutoEncoder.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "AuOkBUP06paA",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "d638924a-9000-47ba-d017-3f42fd385500"
      },
      "cell_type": "code",
      "source": [
        "from keras.layers import Input, Dense, Conv2D, MaxPooling2D, UpSampling2D\n",
        "from keras.models import Model\n",
        "from keras import backend as K\n",
        "\n",
        "input_img = Input(shape=(28, 28, 1))  # adapt this if using `channels_first` image data format\n",
        "\n",
        "x = Conv2D(16, (3, 3), activation='relu', padding='same')(input_img)\n",
        "x = MaxPooling2D((2, 2), padding='same')(x)\n",
        "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
        "x = MaxPooling2D((2, 2), padding='same')(x)\n",
        "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
        "encoded = MaxPooling2D((2, 2), padding='same')(x)\n",
        "\n",
        "# at this point the representation is (4, 4, 8) i.e. 128-dimensional\n",
        "\n",
        "x = Conv2D(8, (3, 3), activation='relu', padding='same')(encoded)\n",
        "x = UpSampling2D((2, 2))(x)\n",
        "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
        "x = UpSampling2D((2, 2))(x)\n",
        "x = Conv2D(16, (3, 3), activation='relu')(x)\n",
        "x = UpSampling2D((2, 2))(x)\n",
        "decoded = Conv2D(1, (3, 3), activation='sigmoid', padding='same')(x)\n",
        "\n",
        "autoencoder = Model(input_img, decoded)\n",
        "autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "cCnPPizj6qoG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "a75b4881-55d9-45c4-8af9-ce28e83bda23"
      },
      "cell_type": "code",
      "source": [
        "from keras.datasets import mnist\n",
        "import numpy as np\n",
        "\n",
        "(x_train, _), (x_test, _) = mnist.load_data()\n",
        "\n",
        "x_train = x_train.astype('float32') / 255.\n",
        "x_test = x_test.astype('float32') / 255.\n",
        "x_train = np.reshape(x_train, (len(x_train), 28, 28, 1))  # adapt this if using `channels_first` image data format\n",
        "x_test = np.reshape(x_test, (len(x_test), 28, 28, 1))  # adapt this if using `channels_first` image data format"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 1s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "AvPqUrcb6xOo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1751
        },
        "outputId": "7cf5ab79-b083-49a8-eec4-422d322d146b"
      },
      "cell_type": "code",
      "source": [
        "from keras.callbacks import TensorBoard\n",
        "\n",
        "autoencoder.fit(x_train, x_train,\n",
        "                epochs=50,\n",
        "                batch_size=128,\n",
        "                shuffle=True,\n",
        "                validation_data=(x_test, x_test))"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 60000 samples, validate on 10000 samples\n",
            "Epoch 1/50\n",
            "60000/60000 [==============================] - 7s 121us/step - loss: 0.1211 - val_loss: 0.1158\n",
            "Epoch 2/50\n",
            "60000/60000 [==============================] - 7s 117us/step - loss: 0.1187 - val_loss: 0.1149\n",
            "Epoch 3/50\n",
            "60000/60000 [==============================] - 7s 113us/step - loss: 0.1163 - val_loss: 0.1148\n",
            "Epoch 4/50\n",
            "60000/60000 [==============================] - 7s 112us/step - loss: 0.1148 - val_loss: 0.1159\n",
            "Epoch 5/50\n",
            "60000/60000 [==============================] - 7s 112us/step - loss: 0.1131 - val_loss: 0.1096\n",
            "Epoch 6/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1120 - val_loss: 0.1095\n",
            "Epoch 7/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1107 - val_loss: 0.1118\n",
            "Epoch 8/50\n",
            "60000/60000 [==============================] - 7s 112us/step - loss: 0.1095 - val_loss: 0.1109\n",
            "Epoch 9/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1087 - val_loss: 0.1077\n",
            "Epoch 10/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1076 - val_loss: 0.1048\n",
            "Epoch 11/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1067 - val_loss: 0.1062\n",
            "Epoch 12/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1064 - val_loss: 0.1069\n",
            "Epoch 13/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1059 - val_loss: 0.1070\n",
            "Epoch 14/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1054 - val_loss: 0.1043\n",
            "Epoch 15/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1052 - val_loss: 0.1039\n",
            "Epoch 16/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1046 - val_loss: 0.1040\n",
            "Epoch 17/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1041 - val_loss: 0.1055\n",
            "Epoch 18/50\n",
            "60000/60000 [==============================] - 7s 112us/step - loss: 0.1036 - val_loss: 0.1017\n",
            "Epoch 19/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1033 - val_loss: 0.1080\n",
            "Epoch 20/50\n",
            "60000/60000 [==============================] - 8s 135us/step - loss: 0.1029 - val_loss: 0.1011\n",
            "Epoch 21/50\n",
            "60000/60000 [==============================] - 9s 143us/step - loss: 0.1026 - val_loss: 0.1009\n",
            "Epoch 22/50\n",
            "60000/60000 [==============================] - 7s 121us/step - loss: 0.1024 - val_loss: 0.1012\n",
            "Epoch 23/50\n",
            "60000/60000 [==============================] - 7s 119us/step - loss: 0.1022 - val_loss: 0.1005\n",
            "Epoch 24/50\n",
            "60000/60000 [==============================] - 7s 113us/step - loss: 0.1019 - val_loss: 0.1001\n",
            "Epoch 25/50\n",
            "60000/60000 [==============================] - 7s 114us/step - loss: 0.1015 - val_loss: 0.0998\n",
            "Epoch 26/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1014 - val_loss: 0.1032\n",
            "Epoch 27/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.1011 - val_loss: 0.1006\n",
            "Epoch 28/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.1010 - val_loss: 0.0982\n",
            "Epoch 29/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1008 - val_loss: 0.0999\n",
            "Epoch 30/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1004 - val_loss: 0.0985\n",
            "Epoch 31/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.1001 - val_loss: 0.0979\n",
            "Epoch 32/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.1000 - val_loss: 0.0978\n",
            "Epoch 33/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0999 - val_loss: 0.0991\n",
            "Epoch 34/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0997 - val_loss: 0.1015\n",
            "Epoch 35/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0995 - val_loss: 0.0978\n",
            "Epoch 36/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.0994 - val_loss: 0.0972\n",
            "Epoch 37/50\n",
            "60000/60000 [==============================] - 7s 113us/step - loss: 0.0992 - val_loss: 0.0993\n",
            "Epoch 38/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.0990 - val_loss: 0.0968\n",
            "Epoch 39/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.0987 - val_loss: 0.0993\n",
            "Epoch 40/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0987 - val_loss: 0.0973\n",
            "Epoch 41/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.0986 - val_loss: 0.0985\n",
            "Epoch 42/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.0984 - val_loss: 0.0968\n",
            "Epoch 43/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0980 - val_loss: 0.0964\n",
            "Epoch 44/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0980 - val_loss: 0.0974\n",
            "Epoch 45/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0979 - val_loss: 0.0983\n",
            "Epoch 46/50\n",
            "60000/60000 [==============================] - 7s 116us/step - loss: 0.0980 - val_loss: 0.0973\n",
            "Epoch 47/50\n",
            "60000/60000 [==============================] - 7s 113us/step - loss: 0.0978 - val_loss: 0.0969\n",
            "Epoch 48/50\n",
            "60000/60000 [==============================] - 7s 111us/step - loss: 0.0977 - val_loss: 0.0958\n",
            "Epoch 49/50\n",
            "60000/60000 [==============================] - 7s 109us/step - loss: 0.0977 - val_loss: 0.0970\n",
            "Epoch 50/50\n",
            "60000/60000 [==============================] - 7s 110us/step - loss: 0.0974 - val_loss: 0.0949\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f3bf7f6eb70>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "1n-Bkaoz67_1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 244
        },
        "outputId": "2e03aca3-be0b-4b68-9406-6b8467f1567b"
      },
      "cell_type": "code",
      "source": [
        "decoded_imgs = autoencoder.predict(x_test)\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "n = 10\n",
        "plt.figure(figsize=(20, 4))\n",
        "for i in range(1, n):\n",
        "    # display original\n",
        "    ax = plt.subplot(2, n, i)\n",
        "    plt.imshow(x_test[i].reshape(28, 28))\n",
        "    plt.gray()\n",
        "    ax.get_xaxis().set_visible(False)\n",
        "    ax.get_yaxis().set_visible(False)\n",
        "\n",
        "    # display reconstruction\n",
        "    ax = plt.subplot(2, n, i + n)\n",
        "    plt.imshow(decoded_imgs[i].reshape(28, 28))\n",
        "    plt.gray()\n",
        "    ax.get_xaxis().set_visible(False)\n",
        "    ax.get_yaxis().set_visible(False)\n",
        "plt.show()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/gAAADjCAYAAAAxIr9SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f3bf7f7b128>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}