02_CNN_3_layers.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bf6a6kIl_rsU"
      },
      "source": [
        "## Import Relevant Libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aQw4TdHC8ba8"
      },
      "outputs": [],
      "source": [
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "import os\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import Input, Conv2D, BatchNormalization, MaxPooling2D, Dropout, GlobalAveragePooling2D, Dense\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import numpy as np\n",
        "import seaborn as sns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "_hX4C6pvThgW"
      },
      "outputs": [],
      "source": [
        "old_folder_name = 'OCT2017 '\n",
        "new_folder_name = 'OCT2017'\n",
        "os.rename(old_folder_name, new_folder_name)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6MadNx_6Aaq-"
      },
      "source": [
        "## Apply CNN Model with 3 layers without Image Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "15P7MVYNAZ_p",
        "outputId": "48dad646-b635-4e42-f738-87e978dd655b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 66788 images belonging to 4 classes.\n",
            "Found 16696 images belonging to 4 classes.\n"
          ]
        }
      ],
      "source": [
        "# Generate batches of tensor image data with real-time data augmentation\n",
        "datagen = ImageDataGenerator(validation_split=0.2)\n",
        "\n",
        "train_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/train\",\n",
        "    target_size = (224, 224),\n",
        "    batch_size=128,\n",
        "    class_mode=\"categorical\",\n",
        "    subset = \"training\"\n",
        ")\n",
        "\n",
        "validation_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/train\",\n",
        "    target_size = (224, 224),\n",
        "    batch_size=128,\n",
        "    class_mode = \"categorical\",\n",
        "    subset = \"validation\"\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-KogYUJMORZo"
      },
      "outputs": [],
      "source": [
        "tf.keras.backend.clear_session()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FlNU-t-bOoOq"
      },
      "outputs": [],
      "source": [
        "# Reference: https://keras.io/api/layers/convolution_layers/convolution2d/\n",
        "# Reference: https://keras.io/api/layers/normalization_layers/batch_normalization/\n",
        "# Reference: https://keras.io/api/layers/pooling_layers/max_pooling2d/\n",
        "# Reference: https://keras.io/api/layers/regularization_layers/dropout/\n",
        "# Reference: https://keras.io/api/layers/pooling_layers/global_average_pooling2d/\n",
        "# Reference: https://keras.io/api/layers/core_layers/dense/\n",
        "\n",
        "input_shape = (224, 224, 3)\n",
        "\n",
        "Input_Layer = Input(shape=(input_shape))\n",
        "\n",
        "# Convolutional layer\n",
        "first_conv2D = Conv2D(32, kernel_size=(3, 3), activation='relu', padding='same')(Input_Layer)\n",
        "# Batch Normalization\n",
        "first_batch_normalization = BatchNormalization()(first_conv2D)\n",
        "\n",
        "# Convolutional layer\n",
        "second_conv2D = Conv2D(64, kernel_size=(3,3), activation='relu', padding='same')(first_batch_normalization)\n",
        "# Batch Normalization\n",
        "second_batch_normalization = BatchNormalization()(second_conv2D)\n",
        "\n",
        "# MaxPool Layer\n",
        "maxpool = MaxPooling2D(pool_size=(2,2))(second_batch_normalization)\n",
        "# Dropout Layer\n",
        "dropout = Dropout(rate=0.2)(maxpool)\n",
        "\n",
        "# Convolutional layer\n",
        "third_conv2D = Conv2D(64, kernel_size=(3,3), activation='relu', padding='same')(dropout)\n",
        "third_batch_normalization = BatchNormalization()(third_conv2D)\n",
        "\n",
        "# GlovalAveragePooling Layer\n",
        "global_average_pool = GlobalAveragePooling2D()(third_batch_normalization)\n",
        "\n",
        "# Dense Layer\n",
        "output_layer = Dense(4, activation=\"softmax\")(global_average_pool)\n",
        "\n",
        "model1 = Model(inputs = Input_Layer, outputs=output_layer)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CtmwzmG6RS_j",
        "outputId": "cfc44400-95d3-435e-aaca-a490e3f28f53"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_5 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " conv2d_8 (Conv2D)           (None, 224, 224, 32)      896       \n",
            "                                                                 \n",
            " batch_normalization_7 (Batc  (None, 224, 224, 32)     128       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " conv2d_9 (Conv2D)           (None, 224, 224, 64)      18496     \n",
            "                                                                 \n",
            " batch_normalization_8 (Batc  (None, 224, 224, 64)     256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 112, 112, 64)     0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 112, 112, 64)      0         \n",
            "                                                                 \n",
            " conv2d_10 (Conv2D)          (None, 112, 112, 64)      36928     \n",
            "                                                                 \n",
            " batch_normalization_9 (Batc  (None, 112, 112, 64)     256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " global_average_pooling2d_2   (None, 64)               0         \n",
            " (GlobalAveragePooling2D)                                        \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 4)                 260       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 57,220\n",
            "Trainable params: 56,900\n",
            "Non-trainable params: 320\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model1.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dya2nYqbRWEt",
        "outputId": "7d69597c-ffc6-410f-c718-6c5b1ad1a553"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.7900 - accuracy: 0.7058\n",
            "Epoch 1: val_accuracy improved from -inf to 0.70053, saving model to model1_without_augmentation_save/weights-01-0.7005.hdf5\n",
            "521/521 [==============================] - 675s 1s/step - loss: 0.7900 - accuracy: 0.7058 - val_loss: 0.8375 - val_accuracy: 0.7005\n",
            "Epoch 2/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.5570 - accuracy: 0.7970\n",
            "Epoch 2: val_accuracy improved from 0.70053 to 0.70688, saving model to model1_without_augmentation_save/weights-02-0.7069.hdf5\n",
            "521/521 [==============================] - 642s 1s/step - loss: 0.5570 - accuracy: 0.7970 - val_loss: 0.7454 - val_accuracy: 0.7069\n",
            "Epoch 3/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.4268 - accuracy: 0.8489\n",
            "Epoch 3: val_accuracy improved from 0.70688 to 0.74784, saving model to model1_without_augmentation_save/weights-03-0.7478.hdf5\n",
            "521/521 [==============================] - 637s 1s/step - loss: 0.4268 - accuracy: 0.8489 - val_loss: 0.6832 - val_accuracy: 0.7478\n",
            "Epoch 4/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.3453 - accuracy: 0.8807\n",
            "Epoch 4: val_accuracy improved from 0.74784 to 0.82577, saving model to model1_without_augmentation_save/weights-04-0.8258.hdf5\n",
            "521/521 [==============================] - 626s 1s/step - loss: 0.3453 - accuracy: 0.8807 - val_loss: 0.5178 - val_accuracy: 0.8258\n",
            "Epoch 5/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.3007 - accuracy: 0.8958\n",
            "Epoch 5: val_accuracy improved from 0.82577 to 0.84146, saving model to model1_without_augmentation_save/weights-05-0.8415.hdf5\n",
            "521/521 [==============================] - 624s 1s/step - loss: 0.3007 - accuracy: 0.8958 - val_loss: 0.4455 - val_accuracy: 0.8415\n",
            "Epoch 6/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2729 - accuracy: 0.9048\n",
            "Epoch 6: val_accuracy did not improve from 0.84146\n",
            "521/521 [==============================] - 614s 1s/step - loss: 0.2729 - accuracy: 0.9048 - val_loss: 0.4727 - val_accuracy: 0.8341\n",
            "Epoch 7/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2547 - accuracy: 0.9112\n",
            "Epoch 7: val_accuracy improved from 0.84146 to 0.88356, saving model to model1_without_augmentation_save/weights-07-0.8836.hdf5\n",
            "521/521 [==============================] - 621s 1s/step - loss: 0.2547 - accuracy: 0.9112 - val_loss: 0.3254 - val_accuracy: 0.8836\n",
            "Epoch 8/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2411 - accuracy: 0.9169\n",
            "Epoch 8: val_accuracy improved from 0.88356 to 0.90381, saving model to model1_without_augmentation_save/weights-08-0.9038.hdf5\n",
            "521/521 [==============================] - 609s 1s/step - loss: 0.2411 - accuracy: 0.9169 - val_loss: 0.2789 - val_accuracy: 0.9038\n",
            "Epoch 9/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2324 - accuracy: 0.9189\n",
            "Epoch 9: val_accuracy did not improve from 0.90381\n",
            "521/521 [==============================] - 667s 1s/step - loss: 0.2324 - accuracy: 0.9189 - val_loss: 0.3948 - val_accuracy: 0.8598\n",
            "Epoch 10/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2212 - accuracy: 0.9232\n",
            "Epoch 10: val_accuracy did not improve from 0.90381\n",
            "521/521 [==============================] - 611s 1s/step - loss: 0.2212 - accuracy: 0.9232 - val_loss: 0.5754 - val_accuracy: 0.8552\n",
            "Epoch 11/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2141 - accuracy: 0.9265\n",
            "Epoch 11: val_accuracy did not improve from 0.90381\n",
            "521/521 [==============================] - 669s 1s/step - loss: 0.2141 - accuracy: 0.9265 - val_loss: 0.4332 - val_accuracy: 0.8579\n",
            "Epoch 12/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2074 - accuracy: 0.9277\n",
            "Epoch 12: val_accuracy improved from 0.90381 to 0.91830, saving model to model1_without_augmentation_save/weights-12-0.9183.hdf5\n",
            "521/521 [==============================] - 607s 1s/step - loss: 0.2074 - accuracy: 0.9277 - val_loss: 0.2209 - val_accuracy: 0.9183\n",
            "Epoch 13/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2040 - accuracy: 0.9291\n",
            "Epoch 13: val_accuracy did not improve from 0.91830\n",
            "521/521 [==============================] - 608s 1s/step - loss: 0.2040 - accuracy: 0.9291 - val_loss: 2.7811 - val_accuracy: 0.3935\n",
            "Epoch 14/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.1962 - accuracy: 0.9316\n",
            "Epoch 14: val_accuracy did not improve from 0.91830\n",
            "521/521 [==============================] - 607s 1s/step - loss: 0.1962 - accuracy: 0.9316 - val_loss: 0.2627 - val_accuracy: 0.9043\n",
            "Epoch 15/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.1895 - accuracy: 0.9336\n",
            "Epoch 15: val_accuracy did not improve from 0.91830\n",
            "521/521 [==============================] - 611s 1s/step - loss: 0.1895 - accuracy: 0.9336 - val_loss: 0.3117 - val_accuracy: 0.8861\n",
            "Epoch 16/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.1872 - accuracy: 0.9344\n",
            "Epoch 16: val_accuracy did not improve from 0.91830\n",
            "521/521 [==============================] - 604s 1s/step - loss: 0.1872 - accuracy: 0.9344 - val_loss: 0.4310 - val_accuracy: 0.8471\n",
            "Epoch 17/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.1835 - accuracy: 0.9355\n",
            "Epoch 17: val_accuracy did not improve from 0.91830\n",
            "521/521 [==============================] - 606s 1s/step - loss: 0.1835 - accuracy: 0.9355 - val_loss: 0.3079 - val_accuracy: 0.8916\n",
            "Epoch 17: early stopping\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7b52904f19f0>"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "earlystop = EarlyStopping(monitor='val_accuracy', min_delta=0, patience=5, verbose=1)\n",
        "filepath=\"model1_without_augmentation_save/weights-{epoch:02d}-{val_accuracy:.4f}.hdf5\"\n",
        "checkpoint = ModelCheckpoint(filepath=filepath, monitor='val_accuracy',verbose=1, save_best_only=True, mode='max')\n",
        "\n",
        "# Callbacks\n",
        "optimizer = tf.keras.optimizers.Adam()\n",
        "model1.compile(optimizer=optimizer,loss=\"categorical_crossentropy\",metrics=[\"accuracy\"])\n",
        "\n",
        "# Creating a list with all the callbacks we want\n",
        "callback_list = [earlystop, checkpoint]\n",
        "\n",
        "train_samples = 66788\n",
        "validation_samples = 16696\n",
        "batch_size = 128\n",
        "model1.fit(train_generator, validation_data= validation_generator, steps_per_epoch=train_samples//batch_size, callbacks=callback_list, epochs=50)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 199
        },
        "id": "kjuiByV0-79B",
        "outputId": "a05b30af-faee-41bb-8e9e-902967354857"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7b5301e77250>"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 3000x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plotting training & Validation accuracy values\n",
        "plt.figure(figsize=(30, 5))\n",
        "plt.subplot(121)\n",
        "plt.plot(model1.history.history['accuracy'])\n",
        "plt.plot(model1.history.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.subplot(122)\n",
        "plt.plot(model1.history.history['loss'])\n",
        "plt.plot(model1.history.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3IsZBGa6_kva",
        "outputId": "9ef06d94-3ad6-4c62-b9ca-8a226bf42456"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 968 images belonging to 4 classes.\n",
            "8/8 [==============================] - 7s 874ms/step - loss: 0.1776 - accuracy: 0.9277\n",
            "Accuracy on Test Dataset: 92.76859760284424, Loss on Test Dataset: 0.17764516174793243\n"
          ]
        }
      ],
      "source": [
        "# Accuracy on test dataset\n",
        "\n",
        "test_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/test\",\n",
        "    target_size = (224, 224),\n",
        "    batch_size = 128,\n",
        "    class_mode=\"categorical\",\n",
        "    shuffle=False\n",
        ")\n",
        "test_samples = 968\n",
        "batch_size = 128\n",
        "\n",
        "result = model1.evaluate(test_generator, steps=test_samples//batch_size+1, verbose=1)\n",
        "print(f\"Accuracy on Test Dataset: {result[1]*100}, Loss on Test Dataset: {result[0]}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 489
        },
        "id": "Eu2Hi22sAGHp",
        "outputId": "b35027c4-7ab8-4224-c37b-a5d672e24a60"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "8/8 [==============================] - 5s 573ms/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Confusion Matrix\n",
        "y_pred = model1.predict(test_generator, test_samples//batch_size+1)\n",
        "predicted_values = np.argmax(y_pred, axis=1)\n",
        "\n",
        "confusion = confusion_matrix(test_generator.classes,predicted_values)\n",
        "plt.title(\"3 Layer CNN without Image Augmentation\")\n",
        "sns.heatmap(confusion, annot=True, cmap=\"YlGnBu\")\n",
        "plt.xlabel('Predicted Values')\n",
        "plt.ylabel('Acutal Values')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u5nDh7mfA7Eq"
      },
      "source": [
        "## Observations:\n",
        "- We are getting an accuracy of 0.92 on the test dataset using simple 3 Layers CNN model without Image Augmentation."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B0qrXHsfBy8u"
      },
      "source": [
        "# Apply 3 Layers CNN Model with Image Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "phfpjbOxApyD",
        "outputId": "1da5858a-8d10-4bde-8acb-f26ae701ed80"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 66788 images belonging to 4 classes.\n",
            "Found 16696 images belonging to 4 classes.\n"
          ]
        }
      ],
      "source": [
        "# generate batches of tensor image data with real-time data augmentation\n",
        "\n",
        "datagen = ImageDataGenerator(rescale=1./255,\n",
        "                             shear_range=0.2,\n",
        "                             zoom_range=0.2,\n",
        "                             horizontal_flip=True,\n",
        "                             validation_split=0.2)\n",
        "\n",
        "train_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/train\",\n",
        "    target_size = (224,224),\n",
        "    batch_size = 128,\n",
        "    class_mode = 'categorical',\n",
        "    subset=\"training\"\n",
        ")\n",
        "\n",
        "validation_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/train\",\n",
        "    target_size=(224, 224),\n",
        "    batch_size = 128,\n",
        "    class_mode = \"categorical\",\n",
        "    subset = \"validation\"\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "axEpvwnbChK7"
      },
      "outputs": [],
      "source": [
        "tf.keras.backend.clear_session()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "dz6Ec0jDCm6x"
      },
      "outputs": [],
      "source": [
        "input_shape = (224,224,3)\n",
        "\n",
        "# Input Layer\n",
        "Input_Layer = Input(shape=(input_shape))\n",
        "\n",
        "# Convolutional Layer\n",
        "First_Conv2D = Conv2D(32, kernel_size = (3, 3), activation=\"relu\",padding = 'same')(Input_Layer)\n",
        "# Batch Normalization\n",
        "First_BatchNormalization = BatchNormalization()(First_Conv2D)\n",
        "\n",
        "# Convolutional Layer\n",
        "Second_Conv2D = Conv2D(64, kernel_size = (3, 3), activation=\"relu\",padding = 'same')(First_BatchNormalization)\n",
        "# Batch Normalization\n",
        "Second_BatchNormalization = BatchNormalization()(Second_Conv2D)\n",
        "\n",
        "# MaxPool Layer\n",
        "maxpool = MaxPooling2D(pool_size=(2,2))(Second_BatchNormalization)\n",
        "# Dropout Layer\n",
        "dropout = Dropout(rate=0.2)(maxpool)\n",
        "\n",
        "# Convolutional Layer\n",
        "Third_Conv2D = Conv2D(64, kernel_size = (3, 3), activation=\"relu\",padding = 'same')(dropout)\n",
        "# Batch Normalization\n",
        "Third_BatchNormalization = BatchNormalization()(Third_Conv2D)\n",
        "\n",
        "# GlobalAveragePooling Layer\n",
        "global_average_pool = GlobalAveragePooling2D()(Third_BatchNormalization)\n",
        "\n",
        "# Dense Layer\n",
        "Output_Layer = Dense(4, activation=\"softmax\")(global_average_pool)\n",
        "\n",
        "# Creating a model\n",
        "model2 = Model(inputs=Input_Layer,outputs=Output_Layer)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ymY0JhzGC_Ub",
        "outputId": "512e8c1c-7afd-4301-f3f6-4532662600bf"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " conv2d (Conv2D)             (None, 224, 224, 32)      896       \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 224, 224, 32)     128       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 224, 224, 64)      18496     \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 224, 224, 64)     256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 112, 112, 64)     0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 112, 112, 64)      0         \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 112, 112, 64)      36928     \n",
            "                                                                 \n",
            " batch_normalization_2 (Batc  (None, 112, 112, 64)     256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " global_average_pooling2d (G  (None, 64)               0         \n",
            " lobalAveragePooling2D)                                          \n",
            "                                                                 \n",
            " dense (Dense)               (None, 4)                 260       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 57,220\n",
            "Trainable params: 56,900\n",
            "Non-trainable params: 320\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model2.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U1SphtupDApI",
        "outputId": "411108a4-9d91-4d23-c3dc-db18afa6b807"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.8685 - accuracy: 0.6739\n",
            "Epoch 1: val_accuracy improved from -inf to 0.44568, saving model to model2_with_augmentation_save/weights-01-0.4457.hdf5\n",
            "521/521 [==============================] - 1471s 3s/step - loss: 0.8685 - accuracy: 0.6739 - val_loss: 5.6457 - val_accuracy: 0.4457\n",
            "Epoch 2/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.6197 - accuracy: 0.7739\n",
            "Epoch 2: val_accuracy improved from 0.44568 to 0.74365, saving model to model2_with_augmentation_save/weights-02-0.7437.hdf5\n",
            "521/521 [==============================] - 1435s 3s/step - loss: 0.6197 - accuracy: 0.7739 - val_loss: 0.6862 - val_accuracy: 0.7437\n",
            "Epoch 3/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.4887 - accuracy: 0.8274\n",
            "Epoch 3: val_accuracy improved from 0.74365 to 0.78360, saving model to model2_with_augmentation_save/weights-03-0.7836.hdf5\n",
            "521/521 [==============================] - 1401s 3s/step - loss: 0.4887 - accuracy: 0.8274 - val_loss: 0.6040 - val_accuracy: 0.7836\n",
            "Epoch 4/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.3996 - accuracy: 0.8607\n",
            "Epoch 4: val_accuracy improved from 0.78360 to 0.82565, saving model to model2_with_augmentation_save/weights-04-0.8256.hdf5\n",
            "521/521 [==============================] - 1391s 3s/step - loss: 0.3996 - accuracy: 0.8607 - val_loss: 0.4629 - val_accuracy: 0.8256\n",
            "Epoch 5/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.3454 - accuracy: 0.8811\n",
            "Epoch 5: val_accuracy did not improve from 0.82565\n",
            "521/521 [==============================] - 1395s 3s/step - loss: 0.3454 - accuracy: 0.8811 - val_loss: 0.4726 - val_accuracy: 0.8226\n",
            "Epoch 6/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.3151 - accuracy: 0.8897\n",
            "Epoch 6: val_accuracy improved from 0.82565 to 0.89632, saving model to model2_with_augmentation_save/weights-06-0.8963.hdf5\n",
            "521/521 [==============================] - 1379s 3s/step - loss: 0.3151 - accuracy: 0.8897 - val_loss: 0.3066 - val_accuracy: 0.8963\n",
            "Epoch 7/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2942 - accuracy: 0.8968\n",
            "Epoch 7: val_accuracy did not improve from 0.89632\n",
            "521/521 [==============================] - 1391s 3s/step - loss: 0.2942 - accuracy: 0.8968 - val_loss: 0.3906 - val_accuracy: 0.8624\n",
            "Epoch 8/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2803 - accuracy: 0.9018\n",
            "Epoch 8: val_accuracy did not improve from 0.89632\n",
            "521/521 [==============================] - 1379s 3s/step - loss: 0.2803 - accuracy: 0.9018 - val_loss: 0.5411 - val_accuracy: 0.7725\n",
            "Epoch 9/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2709 - accuracy: 0.9051\n",
            "Epoch 9: val_accuracy did not improve from 0.89632\n",
            "521/521 [==============================] - 1374s 3s/step - loss: 0.2709 - accuracy: 0.9051 - val_loss: 0.7545 - val_accuracy: 0.7690\n",
            "Epoch 10/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2617 - accuracy: 0.9096\n",
            "Epoch 10: val_accuracy did not improve from 0.89632\n",
            "521/521 [==============================] - 1387s 3s/step - loss: 0.2617 - accuracy: 0.9096 - val_loss: 0.9212 - val_accuracy: 0.6001\n",
            "Epoch 11/50\n",
            "521/521 [==============================] - ETA: 0s - loss: 0.2538 - accuracy: 0.9121\n",
            "Epoch 11: val_accuracy did not improve from 0.89632\n",
            "521/521 [==============================] - 1399s 3s/step - loss: 0.2538 - accuracy: 0.9121 - val_loss: 0.7049 - val_accuracy: 0.7046\n",
            "Epoch 11: early stopping\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x79e7461d3b80>"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "earlystop = EarlyStopping(monitor='val_accuracy', min_delta=0, patience=5, verbose=1)\n",
        "\n",
        "filepath = \"model2_with_augmentation_save/weights-{epoch:02d}-{val_accuracy:.4f}.hdf5\"\n",
        "checkpoint = ModelCheckpoint(filepath=filepath, monitor='val_accuracy', verbose=1, save_best_only=True, mode='max')\n",
        "\n",
        "# callbacks\n",
        "optimizer = tf.keras.optimizers.Adam()\n",
        "model2.compile(optimizer=optimizer, loss=\"categorical_crossentropy\", metrics=['accuracy'])\n",
        "\n",
        "# Creating a list with all the callbacks we want\n",
        "callback_list = [earlystop, checkpoint]\n",
        "\n",
        "train_samples = 66788\n",
        "validation_samples = 16696\n",
        "batch_size = 128\n",
        "model2.fit(train_generator, validation_data=validation_generator, steps_per_epoch=train_samples//batch_size,\n",
        "           callbacks=callback_list,\n",
        "           epochs=50)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        },
        "id": "9WBG8gL8KO5c",
        "outputId": "6b98507f-6305-42be-89cf-f299df2280ad"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x79e7461b7820>"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 3000x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plotting training & Validation accuracy values\n",
        "plt.figure(figsize=(30, 5))\n",
        "plt.subplot(121)\n",
        "plt.plot(model2.history.history['accuracy'])\n",
        "plt.plot(model2.history.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.subplot(122)\n",
        "plt.plot(model2.history.history['loss'])\n",
        "plt.plot(model2.history.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A4PqFnTZKVUT",
        "outputId": "72bc49d6-4603-4c5b-cf7c-bddaaf41524b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Found 968 images belonging to 4 classes.\n",
            "8/8 [==============================] - 12s 1s/step - loss: 0.8291 - accuracy: 0.6560\n",
            "Accuracy on Test Dataset: 65.59917330741882, Loss on Test Dataset: 0.8290548324584961\n"
          ]
        }
      ],
      "source": [
        "# Accuracy on test dataset\n",
        "\n",
        "test_generator = datagen.flow_from_directory(\n",
        "    \"OCT2017/test\",\n",
        "    target_size = (224, 224),\n",
        "    batch_size = 128,\n",
        "    class_mode=\"categorical\",\n",
        "    shuffle=False\n",
        ")\n",
        "test_samples = 968\n",
        "batch_size = 128\n",
        "\n",
        "result = model2.evaluate(test_generator, steps=test_samples//batch_size+1, verbose=1)\n",
        "print(f\"Accuracy on Test Dataset: {result[1]*100}, Loss on Test Dataset: {result[0]}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 489
        },
        "id": "9SaqlRagKZeE",
        "outputId": "b48d787f-3923-4b3b-cddf-05de765ddeae"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "8/8 [==============================] - 13s 2s/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Confusion Matrix\n",
        "y_pred = model2.predict(test_generator, test_samples//batch_size+1)\n",
        "predicted_values = np.argmax(y_pred, axis=1)\n",
        "\n",
        "confusion = confusion_matrix(test_generator.classes,predicted_values)\n",
        "plt.title(\"3 Layer CNN with Image Augmentation\")\n",
        "sns.heatmap(confusion, annot=True, cmap=\"YlGnBu\")\n",
        "plt.xlabel('Predicted Values')\n",
        "plt.ylabel('Acutal Values')\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}