FASHIONMNIST_Autoencoders.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z9CDghVZ4W5m"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9Zp6CCzeGQR5",
        "outputId": "877f5b89-99c9-473f-e94c-fae749d1bf06"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1JJ5_siReRGH"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BfN5BG5PGe3w",
        "outputId": "1db9eff9-964d-45ca-e5af-538279003b8b"
      },
      "source": [
        "!pip install keras"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: keras in /usr/local/lib/python3.7/dist-packages (2.6.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kMkSM533Gcdi"
      },
      "source": [
        "import keras\n",
        "from matplotlib import pyplot as plt\n",
        "import numpy as np\n",
        "import gzip\n",
        "%matplotlib inline\n",
        "from keras.models import Model\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "# from keras.optimizers import RMSprop\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "ea2O0H7AHVqk"
      },
      "source": [
        "import os\n",
        "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\"\n",
        "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P2wllSnyG9qz"
      },
      "source": [
        "# from keras.layers import Input,Dense,Flatten,Dropout,merge,Reshape,Conv2D,MaxPooling2D,UpSampling2D,Conv2DTranspose\n",
        "# from keras.layers.normalization import BatchNormalization\n",
        "\n",
        "from tensorflow.keras.layers import (\n",
        "    BatchNormalization, Input,Dense,Flatten,Dropout,Reshape,Conv2D,MaxPooling2D,UpSampling2D,Conv2DTranspose\n",
        ")\n",
        "\n",
        "\n",
        "from keras.models import Model,Sequential\n",
        "from keras.callbacks import ModelCheckpoint\n",
        "from tensorflow.keras.optimizers import Adadelta, RMSprop,SGD,Adam\n",
        "from keras import regularizers\n",
        "from keras import backend as K\n",
        "# from keras.utils import to_categorical"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AConiStlGUgM"
      },
      "source": [
        "def extract_data(filename, num_images):\n",
        "    with gzip.open(filename) as bytestream:\n",
        "        bytestream.read(16)\n",
        "        buf = bytestream.read(28 * 28 * num_images)\n",
        "        data = np.frombuffer(buf, dtype=np.uint8).astype(np.float32)\n",
        "        data = data.reshape(num_images, 28,28)\n",
        "        return data\n",
        "\n",
        "def extract_labels(filename, num_images):\n",
        "    with gzip.open(filename) as bytestream:\n",
        "        bytestream.read(8)\n",
        "        buf = bytestream.read(1 * num_images)\n",
        "        labels = np.frombuffer(buf, dtype=np.uint8).astype(np.int64)\n",
        "        return labels"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RaUY7Xv7GUoB"
      },
      "source": [
        "train_data = extract_data('/content/drive/MyDrive/7th- Semester/FashionMNIST/train-images-idx3-ubyte.gz', 60000)\n",
        "test_data = extract_data('/content/drive/MyDrive/7th- Semester/FashionMNIST/t10k-images-idx3-ubyte.gz', 10000)\n",
        "\n",
        "\n",
        "train_labels = extract_labels('/content/drive/MyDrive/7th- Semester/FashionMNIST/train-labels-idx1-ubyte.gz',60000)\n",
        "test_labels = extract_labels('/content/drive/MyDrive/7th- Semester/FashionMNIST/t10k-labels-idx1-ubyte.gz',10000)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z56xXc14GUq0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "53620bb0-f979-4641-d287-c54be3e2a44b"
      },
      "source": [
        "print(\"Training set (images) shape: {shape}\".format(shape=train_data.shape))\n",
        "\n",
        "# Shapes of test set\n",
        "print(\"Test set (images) shape: {shape}\".format(shape=test_data.shape))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training set (images) shape: (60000, 28, 28)\n",
            "Test set (images) shape: (10000, 28, 28)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OAOJrj_QGUtp"
      },
      "source": [
        "# Create dictionary of target classes\n",
        "label_dict = {\n",
        " 0: 'A',\n",
        " 1: 'B',\n",
        " 2: 'C',\n",
        " 3: 'D',\n",
        " 4: 'E',\n",
        " 5: 'F',\n",
        " 6: 'G',\n",
        " 7: 'H',\n",
        " 8: 'I',\n",
        " 9: 'J',\n",
        "}"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 208
        },
        "id": "nP8mcN5tGUwd",
        "outputId": "71d65098-eacd-43e9-d654-25f5effb451d"
      },
      "source": [
        "plt.figure(figsize=[5,5])\n",
        "\n",
        "# Display the first image in training data\n",
        "plt.subplot(121)\n",
        "curr_img = np.reshape(train_data[10], (28,28))\n",
        "curr_lbl = train_labels[10]\n",
        "plt.imshow(curr_img, cmap='gray')\n",
        "plt.title(\"(Label: \" + str(label_dict[curr_lbl]) + \")\")\n",
        "\n",
        "# Display the first image in testing data\n",
        "plt.subplot(122)\n",
        "curr_img = np.reshape(test_data[10], (28,28))\n",
        "curr_lbl = test_labels[10]\n",
        "plt.imshow(curr_img, cmap='gray')\n",
        "plt.title(\"(Label: \" + str(label_dict[curr_lbl]) + \")\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, '(Label: E)')"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dWMrXvW_GUzH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "00a5e060-5acb-4abc-b861-a92813f2638b"
      },
      "source": [
        "train_data = train_data.reshape(-1, 28,28, 1)\n",
        "test_data = test_data.reshape(-1, 28,28, 1)\n",
        "train_data.shape, test_data.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((60000, 28, 28, 1), (10000, 28, 28, 1))"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nHDcJk3sGU1q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d240d7e6-e78d-4cee-e9e1-47ad4f096c3e"
      },
      "source": [
        "train_data.dtype, test_data.dtype"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(dtype('float32'), dtype('float32'))"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lQImeBXquwir",
        "outputId": "b7d76d4a-01f7-4884-9fc7-6d098bff49e0"
      },
      "source": [
        "np.max(train_data), np.max(test_data)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(255.0, 255.0)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GGU0eftmuwda"
      },
      "source": [
        "train_data = train_data / np.max(train_data)\n",
        "test_data = test_data / np.max(test_data)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "arLMnyBruwVJ",
        "outputId": "115c9b97-c98c-4092-e122-d64b689598fb"
      },
      "source": [
        "np.max(train_data), np.max(test_data)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1.0, 1.0)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FLKDjyp8uwSb"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "train_X,valid_X,train_ground,valid_ground = train_test_split(train_data,\n",
        "                                                             train_data,\n",
        "                                                             test_size=0.2,\n",
        "                                                             random_state=13)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rplNNQ7hu8nU"
      },
      "source": [
        "batch_size = 64\n",
        "epochs = 1\n",
        "inChannel = 1\n",
        "x, y = 28, 28\n",
        "input_img = Input(shape = (x, y, inChannel))\n",
        "num_classes = 10"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K0BCbBcRu9Xj"
      },
      "source": [
        "def encoder(input_img):\n",
        "    #encoder\n",
        "    #input = 28 x 28 x 1 (wide and thin)\n",
        "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(input_img) #28 x 28 x 32\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv1)\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    pool1 = MaxPooling2D(pool_size=(2, 2))(conv1) #14 x 14 x 32\n",
        "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(pool1) #14 x 14 x 64\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv2)\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    pool2 = MaxPooling2D(pool_size=(2, 2))(conv2) #7 x 7 x 64\n",
        "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(pool2) #7 x 7 x 128 (small and thick)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv3)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv3) #7 x 7 x 256 (small and thick)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv4)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    return conv4\n",
        "\n",
        "def decoder(conv4):    \n",
        "    #decoder\n",
        "    conv5 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv4) #7 x 7 x 128\n",
        "    conv5 = BatchNormalization()(conv5)\n",
        "    conv5 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv5)\n",
        "    conv5 = BatchNormalization()(conv5)\n",
        "    conv6 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv5) #7 x 7 x 64\n",
        "    conv6 = BatchNormalization()(conv6)\n",
        "    conv6 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv6)\n",
        "    conv6 = BatchNormalization()(conv6)\n",
        "    up1 = UpSampling2D((2,2))(conv6) #14 x 14 x 64\n",
        "    conv7 = Conv2D(32, (3, 3), activation='relu', padding='same')(up1) # 14 x 14 x 32\n",
        "    conv7 = BatchNormalization()(conv7)\n",
        "    conv7 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv7)\n",
        "    conv7 = BatchNormalization()(conv7)\n",
        "    up2 = UpSampling2D((2,2))(conv7) # 28 x 28 x 32\n",
        "    decoded = Conv2D(1, (3, 3), activation='sigmoid', padding='same')(up2) # 28 x 28 x 1\n",
        "    return decoded"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ncerekciu9UL"
      },
      "source": [
        "autoencoder = Model(input_img, decoder(encoder(input_img)))\n",
        "autoencoder.compile(loss='mean_squared_error', optimizer = RMSprop())\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "juCLkze4u9RT",
        "outputId": "c093697d-5378-4ee7-f27c-d94a30ac5a00"
      },
      "source": [
        "autoencoder.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_1 (InputLayer)         [(None, 28, 28, 1)]       0         \n",
            "_________________________________________________________________\n",
            "conv2d (Conv2D)              (None, 28, 28, 32)        320       \n",
            "_________________________________________________________________\n",
            "batch_normalization (BatchNo (None, 28, 28, 32)        128       \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 28, 28, 32)        9248      \n",
            "_________________________________________________________________\n",
            "batch_normalization_1 (Batch (None, 28, 28, 32)        128       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 14, 14, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 14, 14, 64)        18496     \n",
            "_________________________________________________________________\n",
            "batch_normalization_2 (Batch (None, 14, 14, 64)        256       \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 14, 14, 64)        36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_3 (Batch (None, 14, 14, 64)        256       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 7, 7, 64)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 7, 7, 128)         73856     \n",
            "_________________________________________________________________\n",
            "batch_normalization_4 (Batch (None, 7, 7, 128)         512       \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 7, 7, 128)         147584    \n",
            "_________________________________________________________________\n",
            "batch_normalization_5 (Batch (None, 7, 7, 128)         512       \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 7, 7, 256)         295168    \n",
            "_________________________________________________________________\n",
            "batch_normalization_6 (Batch (None, 7, 7, 256)         1024      \n",
            "_________________________________________________________________\n",
            "conv2d_7 (Conv2D)            (None, 7, 7, 256)         590080    \n",
            "_________________________________________________________________\n",
            "batch_normalization_7 (Batch (None, 7, 7, 256)         1024      \n",
            "_________________________________________________________________\n",
            "conv2d_8 (Conv2D)            (None, 7, 7, 128)         295040    \n",
            "_________________________________________________________________\n",
            "batch_normalization_8 (Batch (None, 7, 7, 128)         512       \n",
            "_________________________________________________________________\n",
            "conv2d_9 (Conv2D)            (None, 7, 7, 128)         147584    \n",
            "_________________________________________________________________\n",
            "batch_normalization_9 (Batch (None, 7, 7, 128)         512       \n",
            "_________________________________________________________________\n",
            "conv2d_10 (Conv2D)           (None, 7, 7, 64)          73792     \n",
            "_________________________________________________________________\n",
            "batch_normalization_10 (Batc (None, 7, 7, 64)          256       \n",
            "_________________________________________________________________\n",
            "conv2d_11 (Conv2D)           (None, 7, 7, 64)          36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_11 (Batc (None, 7, 7, 64)          256       \n",
            "_________________________________________________________________\n",
            "up_sampling2d (UpSampling2D) (None, 14, 14, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_12 (Conv2D)           (None, 14, 14, 32)        18464     \n",
            "_________________________________________________________________\n",
            "batch_normalization_12 (Batc (None, 14, 14, 32)        128       \n",
            "_________________________________________________________________\n",
            "conv2d_13 (Conv2D)           (None, 14, 14, 32)        9248      \n",
            "_________________________________________________________________\n",
            "batch_normalization_13 (Batc (None, 14, 14, 32)        128       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_1 (UpSampling2 (None, 28, 28, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_14 (Conv2D)           (None, 28, 28, 1)         289       \n",
            "=================================================================\n",
            "Total params: 1,758,657\n",
            "Trainable params: 1,755,841\n",
            "Non-trainable params: 2,816\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ycy80mHDu9NN",
        "outputId": "d5725e29-5bbf-455a-f789-70437d925b65"
      },
      "source": [
        "autoencoder_train = autoencoder.fit(train_X, train_ground, batch_size=batch_size,epochs=epochs,verbose=1,validation_data=(valid_X, valid_ground))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "750/750 [==============================] - 944s 1s/step - loss: 0.0196 - val_loss: 0.0113\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3Lfx07w8u9Ka",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "outputId": "e0d542d0-51ed-4c28-9062-561db42941ec"
      },
      "source": [
        "loss = autoencoder_train.history['loss']\n",
        "val_loss = autoencoder_train.history['val_loss']\n",
        "# epochs = range(200)\n",
        "plt.figure()\n",
        "plt.plot(1, loss, 'bo', label='Training loss')\n",
        "plt.plot(1, val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YB0L0cx5u9Hj"
      },
      "source": [
        "autoencoder.save_weights('/content/drive/MyDrive/7th- Semester/FashionMNIST/FASHIONMNISTautoencoder.h5')\n",
        "# '/content/drive/MyDrive/7th- Semester/FashionMNIST/t10k-labels-idx1-ubyte.gz',10000)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pqx6c_9rQSi4"
      },
      "source": [
        "import keras.utils\n",
        "from keras import utils as np_utils"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JP4uKjjcQSlz",
        "outputId": "aeaa3f05-adb0-4559-c0b1-ea6a2b99db70"
      },
      "source": [
        "# Segmenting the fashion mnist images\n",
        "# Now you will be using the trained autoencoder's head, i.e., the encoder part and will be loading the weights of the autoencoder you just now trained but only in the encoder part of the model.\n",
        "# You will add a few dense or fully connected layers to the encoder to classify fashion mnist images.\n",
        "#     Let's first convert the labels into one-hot encoding vectors.\n",
        "#     For those of you who do not know about one-hot encoding:\n",
        "#     In one-hot encoding, you convert the categorical data into a vector of numbers. \n",
        "# The reason why you convert the categorical data in one hot encoding is that machine learning algorithms cannot work with categorical data directly.\n",
        "#  You generate one boolean column for each category or class. Only one of these columns could take on the value 1 for each sample. Hence, the term one-hot encoding.\n",
        "#     For your problem statement, the one hot encoding will be a row vector, and for each image,\n",
        "#  it will have a dimension of 1 x 10. The important thing to note here is that the vector consists of all zeros except for the class that it represents, and for that, it is 1.\n",
        "#     So let's convert the labels into one-hot encoding vectors:\n",
        "from tensorflow.keras.utils import to_categorical # Change the labels from categorical to one-hot encoding\n",
        "train_Y_one_hot = to_categorical(train_labels) \n",
        "test_Y_one_hot = to_categorical(test_labels)\n",
        "# Display the change for category label using one-hot encoding\n",
        "print('Original label:', train_labels[0])\n",
        "print('After conversion to one-hot:', train_Y_one_hot[0])\n",
        "\n",
        "# That's pretty clear, right?\n",
        "    # This last step is a crucial one which you all have done before as well which is splitting the data into training and validation in the autoencoder training phase. So let's quickly do that step for classification as well. Note that you will use the same random state that you used before and also this time you will pass your labels as well that you just now converted into a one-hot encoding vector.\n",
        "\n",
        "    \n",
        "    "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original label: 9\n",
            "After conversion to one-hot: [0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nO0qBTEsQ6JG",
        "outputId": "ce68026c-34f4-4e07-e4db-1de4c9b49ece"
      },
      "source": [
        "train_X,valid_X,train_label,valid_label = train_test_split(train_data,train_Y_one_hot,test_size=0.2,random_state=13)\n",
        "# For one last time let's check the shape of training and validation set.\n",
        "train_X.shape,valid_X.shape,train_label.shape,valid_label.shape\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((48000, 28, 28, 1), (12000, 28, 28, 1), (48000, 10), (12000, 10))"
            ]
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dzoh18wfvk1V"
      },
      "source": [
        "# autoencoder.save_weights('autoencoder.h5')\n",
        "\n",
        "def encoder(input_img):\n",
        "    #encoder\n",
        "    #input = 28 x 28 x 1 (wide and thin)\n",
        "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(input_img) #28 x 28 x 32\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv1)\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    pool1 = MaxPooling2D(pool_size=(2, 2))(conv1) #14 x 14 x 32\n",
        "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(pool1) #14 x 14 x 64\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv2)\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    pool2 = MaxPooling2D(pool_size=(2, 2))(conv2) #7 x 7 x 64\n",
        "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(pool2) #7 x 7 x 128 (small and thick)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv3)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv3) #7 x 7 x 256 (small and thick)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv4)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    return conv4\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m4VXpERxKZzk",
        "outputId": "8a0b4182-056f-407e-cace-df6b0e323979"
      },
      "source": [
        "input_img[0][0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<KerasTensor: shape=(28, 1) dtype=float32 (created by layer 'tf.__operators__.getitem_1')>"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XU-mLPxOvktC"
      },
      "source": [
        "def fc(enco):\n",
        "    flat = Flatten()(enco)\n",
        "    den = Dense(128, activation='relu')(flat)\n",
        "    out = Dense(num_classes, activation='softmax')(den)\n",
        "    return out\n",
        "\n",
        "encode = encoder(input_img)\n",
        "full_model = Model(input_img,fc(encode))\n",
        "\n",
        "for l1,l2 in zip(full_model.layers[:19],autoencoder.layers[0:19]):\n",
        "    l1.set_weights(l2.get_weights())"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V8t5RLEKvkht",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fb21981c-013c-43b0-fb00-f97e27160570"
      },
      "source": [
        "autoencoder.get_weights()[0][1]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[-0.10972736,  0.1412479 ,  0.04480466, -0.1497556 ,\n",
              "         -0.09553414,  0.12620619,  0.14871094,  0.02645838,\n",
              "          0.00315898,  0.03999595,  0.12085675, -0.07324249,\n",
              "         -0.05615313,  0.03119145,  0.00192349,  0.06490435,\n",
              "         -0.03936107, -0.06159118, -0.00608666,  0.09949011,\n",
              "         -0.04740502, -0.14852351,  0.10553352,  0.07978274,\n",
              "          0.14853783, -0.08903937,  0.10618052,  0.00050346,\n",
              "          0.05130543,  0.14782082,  0.15119953, -0.10190051]],\n",
              "\n",
              "       [[-0.11406079, -0.11755417,  0.12420368,  0.06677188,\n",
              "          0.10806774,  0.12401075,  0.00315239, -0.11059904,\n",
              "         -0.143833  , -0.03489171,  0.12246311,  0.07534908,\n",
              "         -0.03430961, -0.06675816,  0.04640457, -0.10350453,\n",
              "         -0.114095  , -0.07009452,  0.02028428,  0.02610117,\n",
              "         -0.09709974, -0.13219036, -0.10051478, -0.0036644 ,\n",
              "         -0.00646222,  0.12139625,  0.14589262, -0.07067588,\n",
              "         -0.14373446,  0.07853204,  0.09411597,  0.11094949]],\n",
              "\n",
              "       [[ 0.09897374, -0.07187858, -0.12328333, -0.11921339,\n",
              "         -0.08858172,  0.00853375,  0.16631809,  0.00377833,\n",
              "          0.11250199, -0.07313662, -0.00309738,  0.05333624,\n",
              "          0.01555657,  0.00594863, -0.02654916, -0.1399205 ,\n",
              "          0.07451937,  0.03628719,  0.0364215 , -0.09417249,\n",
              "         -0.06810916,  0.1239612 , -0.13789344, -0.01619685,\n",
              "          0.06060186,  0.11779079,  0.04297927, -0.02135263,\n",
              "         -0.01011473,  0.03362168, -0.04119838,  0.12056082]]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uQah_6ht5NV4"
      },
      "source": [
        "encode = encoder(input_img)\n",
        "full_model = Model(input_img,fc(encode))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UMoW-Zj45Ns0"
      },
      "source": [
        "for l1,l2 in zip(full_model.layers[:19],autoencoder.layers[0:19]):\n",
        "    l1.set_weights(l2.get_weights())"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pC8JgXeA5NwX",
        "outputId": "3a50d443-2503-4add-995f-3b68f9c415e2"
      },
      "source": [
        "full_model.get_weights()[0][1]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[-0.10972736,  0.1412479 ,  0.04480466, -0.1497556 ,\n",
              "         -0.09553414,  0.12620619,  0.14871094,  0.02645838,\n",
              "          0.00315898,  0.03999595,  0.12085675, -0.07324249,\n",
              "         -0.05615313,  0.03119145,  0.00192349,  0.06490435,\n",
              "         -0.03936107, -0.06159118, -0.00608666,  0.09949011,\n",
              "         -0.04740502, -0.14852351,  0.10553352,  0.07978274,\n",
              "          0.14853783, -0.08903937,  0.10618052,  0.00050346,\n",
              "          0.05130543,  0.14782082,  0.15119953, -0.10190051]],\n",
              "\n",
              "       [[-0.11406079, -0.11755417,  0.12420368,  0.06677188,\n",
              "          0.10806774,  0.12401075,  0.00315239, -0.11059904,\n",
              "         -0.143833  , -0.03489171,  0.12246311,  0.07534908,\n",
              "         -0.03430961, -0.06675816,  0.04640457, -0.10350453,\n",
              "         -0.114095  , -0.07009452,  0.02028428,  0.02610117,\n",
              "         -0.09709974, -0.13219036, -0.10051478, -0.0036644 ,\n",
              "         -0.00646222,  0.12139625,  0.14589262, -0.07067588,\n",
              "         -0.14373446,  0.07853204,  0.09411597,  0.11094949]],\n",
              "\n",
              "       [[ 0.09897374, -0.07187858, -0.12328333, -0.11921339,\n",
              "         -0.08858172,  0.00853375,  0.16631809,  0.00377833,\n",
              "          0.11250199, -0.07313662, -0.00309738,  0.05333624,\n",
              "          0.01555657,  0.00594863, -0.02654916, -0.1399205 ,\n",
              "          0.07451937,  0.03628719,  0.0364215 , -0.09417249,\n",
              "         -0.06810916,  0.1239612 , -0.13789344, -0.01619685,\n",
              "          0.06060186,  0.11779079,  0.04297927, -0.02135263,\n",
              "         -0.01011473,  0.03362168, -0.04119838,  0.12056082]]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FZCVJ3Hr5N2O"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UcVh6Kw05N7V"
      },
      "source": [
        "# Voila! Both the arrays look exactly similar. So, without any further ado, let's compile the model and start the training.\n",
        "\n",
        "# Next, you will make the encoder part i.e.the first nineteen layers of the model trainable false. \n",
        "# Since the encoder part is already trained, you do not need to train it. You will only be training the Fully Connected part.\n",
        "\n",
        "for layer in full_model.layers[0:19]:\n",
        "    layer.trainable = False\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zfB8bznL5OXR"
      },
      "source": [
        "full_model.compile(loss=keras.losses.categorical_crossentropy, optimizer=Adam(),metrics=['accuracy'])\n",
        "# train_X,valid_X,train_ground,valid_ground \n",
        "# classify_train = full_model.fit(train_X, train_label, batch_size=64,epochs=100,verbose=1,validation_data=(valid_X, valid_label))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cOCzqFEW93KI",
        "outputId": "d3e2ceb4-3d4d-4380-cf53-ca8f1beb7e83"
      },
      "source": [
        "# classify_train = full_model.fit(train_X, train_ground, batch_size=64,epochs=100,verbose=1,validation_data=(valid_X, valid_ground))\n",
        "classify_train = full_model.fit(train_X, train_label, batch_size=64,epochs=1,verbose=1,validation_data=(valid_X, valid_label))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "750/750 [==============================] - 163s 216ms/step - loss: 0.5300 - accuracy: 0.8377 - val_loss: 0.3886 - val_accuracy: 0.8642\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M_c73PrlKyDI",
        "outputId": "09af068d-40a9-47f3-d70b-cbf69e0db17d"
      },
      "source": [
        "train_X.shape,valid_X.shape,train_label.shape,valid_label.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((48000, 28, 28, 1), (12000, 28, 28, 1), (48000, 10), (12000, 10))"
            ]
          },
          "metadata": {},
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xi1D72fETnFf"
      },
      "source": [
        "full_model.save_weights('/content/drive/MyDrive/7th- Semester/FashionMNIST/FASHIONMNIST_classification_complete.h5')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nmqHLyrHTnc9"
      },
      "source": [
        "accuracy = classify_train.history['acc']\n",
        "val_accuracy = classify_train.history['val_acc']\n",
        "loss = classify_train.history['loss']\n",
        "val_loss = classify_train.history['val_loss']\n",
        "epochs = range(len(accuracy))\n",
        "plt.plot(epochs, accuracy, 'bo', label='Training accuracy')\n",
        "plt.plot(epochs, val_accuracy, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "plt.plot(epochs, loss, 'bo', label='Training loss')\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FpEe8jtzTnUO"
      },
      "source": [
        "test_eval = full_model.evaluate(test_data, test_Y_one_hot, verbose=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bxbpvwgWUrKM",
        "outputId": "fb3936a9-1ba5-438f-c2f4-e5bcfa7309b5"
      },
      "source": [
        "print('Test loss:', test_eval[0])\n",
        "print('Test accuracy:', test_eval[1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test loss: 0.41077715158462524\n",
            "Test accuracy: 0.8565000295639038\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fTSYTfJ5U2ns"
      },
      "source": [
        "predicted_classes = full_model.predict(test_data)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WRPl8uBSU6CT",
        "outputId": "ae0c461f-23be-433f-ccd9-8c3d5d1e0fc4"
      },
      "source": [
        "predicted_classes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[9.68984892e-09, 7.44540596e-10, 2.97807307e-10, ...,\n",
              "        3.72564798e-04, 7.17233206e-09, 9.99621987e-01],\n",
              "       [5.20240108e-04, 1.98145244e-07, 9.98265445e-01, ...,\n",
              "        6.05053854e-12, 4.84370548e-06, 4.22508339e-09],\n",
              "       [4.69442790e-13, 1.00000000e+00, 2.12531697e-16, ...,\n",
              "        6.30839510e-14, 9.78159238e-15, 2.14527725e-20],\n",
              "       ...,\n",
              "       [6.74457333e-05, 2.13229691e-06, 4.27077175e-05, ...,\n",
              "        1.09895346e-07, 9.98318791e-01, 3.14691260e-06],\n",
              "       [2.54944954e-10, 9.99998093e-01, 1.36849669e-08, ...,\n",
              "        1.33550476e-10, 1.34166669e-08, 1.59423880e-11],\n",
              "       [4.90065077e-06, 4.80041024e-07, 3.48181420e-05, ...,\n",
              "        2.93680787e-01, 1.17595878e-03, 4.74074564e-04]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T6AP2dH6VKf4"
      },
      "source": [
        "predicted_classes = np.argmax(np.round(predicted_classes),axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CVpptrVNVQvE",
        "outputId": "d1e4b49f-6759-4f22-d24d-1b45af12d08e"
      },
      "source": [
        "predicted_classes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1, ..., 8, 1, 5])"
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w4R2NTxCViqm",
        "outputId": "8b9a85a2-3803-4c4a-a4cc-d16d7573bcee"
      },
      "source": [
        "predicted_classes.shape, test_labels.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((10000,), (10000,))"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 314
        },
        "id": "AdbpNPatVlUq",
        "outputId": "7ffed7e1-f06f-488a-b12c-6a7832801bbe"
      },
      "source": [
        "correct = np.where(predicted_classes==test_labels)[0]\n",
        "print (\"Found %d correct labels\" % len(correct))\n",
        "for i, correct in enumerate(correct[:9]):\n",
        "    plt.subplot(3,3,i+1)\n",
        "    plt.imshow(test_data[correct].reshape(28,28), cmap='gray', interpolation='none')\n",
        "    plt.title(\"Predicted {}, Class {}\".format(predicted_classes[correct], test_labels[correct]))\n",
        "    plt.tight_layout()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 8444 correct labels\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 9 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v5lh8EN9VoUe",
        "outputId": "d9e7bd37-d71e-4977-90dd-9f2cf45fb707"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "target_names = [\"Class {}\".format(i) for i in range(num_classes)]\n",
        "print(classification_report(test_labels, predicted_classes, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     Class 0       0.58      0.89      0.70      1000\n",
            "     Class 1       0.98      0.96      0.97      1000\n",
            "     Class 2       0.85      0.61      0.71      1000\n",
            "     Class 3       0.88      0.88      0.88      1000\n",
            "     Class 4       0.76      0.79      0.78      1000\n",
            "     Class 5       0.94      0.97      0.95      1000\n",
            "     Class 6       0.72      0.56      0.63      1000\n",
            "     Class 7       0.95      0.89      0.92      1000\n",
            "     Class 8       0.98      0.95      0.96      1000\n",
            "     Class 9       0.93      0.94      0.94      1000\n",
            "\n",
            "    accuracy                           0.84     10000\n",
            "   macro avg       0.86      0.84      0.84     10000\n",
            "weighted avg       0.86      0.84      0.84     10000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cnTpAs_CWCET"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Srkn2gXcWCBK"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DW22xArxWB9U"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YZzQrVOdWB6A"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eqCqlc8oWB14"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    }
  ]
}