{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2e67e032-6ac0-4412-abde-51c1a846cb25",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\n",
    "from tensorflow.keras.layers import Dense, Flatten, Dropout, GlobalAveragePooling2D\n",
    "from tensorflow.keras.applications import ResNet50\n",
    "from tensorflow.keras.applications import EfficientNetB0\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from sklearn.metrics import confusion_matrix, accuracy_score, precision_score, recall_score, f1_score, classification_report\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator, load_img, img_to_array\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import shutil"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "16cf6b0c-7ec4-4795-afcd-b06b8c5dfb14",
   "metadata": {},
   "outputs": [],
   "source": [
    "original_dataset_dir = 'eye dataset'  # Path to the dataset with subfolders\n",
    "base_dir = 'dataset'  # Where you want to save new directories\n",
    "\n",
    "train_dir = os.path.join(base_dir, 'train')\n",
    "validation_dir = os.path.join(base_dir, 'validation')\n",
    "test_dir = os.path.join(base_dir, 'test')\n",
    "\n",
    "# Create directories if they do not exist\n",
    "for dir_path in [train_dir, validation_dir, test_dir]:\n",
    "    os.makedirs(dir_path, exist_ok=True)\n",
    "\n",
    "# Split ratios\n",
    "train_ratio = 0.7\n",
    "validation_ratio = 0.1\n",
    "test_ratio = 0.2\n",
    "\n",
    "# Function to split and copy files\n",
    "def split_and_copy_files(class_dir, dest_base_dir):\n",
    "    all_images = [os.path.join(class_dir, f) for f in os.listdir(class_dir) if os.path.isfile(os.path.join(class_dir, f))]\n",
    "    \n",
    "    train_images, temp_images = train_test_split(all_images, test_size=(1 - train_ratio), random_state=42)\n",
    "    validation_images, test_images = train_test_split(temp_images, test_size=test_ratio/(validation_ratio + test_ratio), random_state=42)\n",
    "    \n",
    "    for images, folder_name in zip([train_images, validation_images, test_images], ['train', 'validation', 'test']):\n",
    "        dest_dir = os.path.join(dest_base_dir, folder_name, os.path.basename(class_dir))\n",
    "        os.makedirs(dest_dir, exist_ok=True)\n",
    "        for file in images:\n",
    "            shutil.copy(file, dest_dir)\n",
    "\n",
    "# Iterate over each class directory and split the files\n",
    "for class_dir in os.listdir(original_dataset_dir):\n",
    "    full_class_dir = os.path.join(original_dataset_dir, class_dir)\n",
    "    if os.path.isdir(full_class_dir):\n",
    "        split_and_copy_files(full_class_dir, base_dir)\n",
    "\n",
    "# Function to create data generators\n",
    "def create_data_generators(train_dir, val_dir, test_dir, img_size=(240, 240), batch_size=32):\n",
    "    # Data augmentation for training dataset\n",
    "    train_datagen = ImageDataGenerator(\n",
    "        rescale=1./255,\n",
    "        rotation_range=360,\n",
    "        horizontal_flip=True,\n",
    "        vertical_flip=True,\n",
    "        brightness_range=(0.75, 1.25)\n",
    "    )\n",
    "    \n",
    "    # Only rescale for validation and testing\n",
    "    val_test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "    \n",
    "    # Generate augmented training images\n",
    "    train_generator = train_datagen.flow_from_directory(\n",
    "        train_dir,\n",
    "        target_size=img_size,\n",
    "        batch_size=batch_size,\n",
    "        class_mode='categorical',\n",
    "        shuffle=True\n",
    "    )\n",
    "    \n",
    "    # Generate validation images\n",
    "    val_generator = val_test_datagen.flow_from_directory(\n",
    "        val_dir,\n",
    "        target_size=img_size,\n",
    "        batch_size=batch_size,\n",
    "        class_mode='categorical',\n",
    "        shuffle=False\n",
    "    )\n",
    "    \n",
    "    # Generate test images\n",
    "    test_generator = val_test_datagen.flow_from_directory(\n",
    "        test_dir,\n",
    "        target_size=img_size,\n",
    "        batch_size=batch_size,\n",
    "        class_mode='categorical',\n",
    "        shuffle=False\n",
    "    )\n",
    "    \n",
    "    return train_generator, val_generator, test_generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ca7fdfaa-b8fc-4e54-9abd-3e12f04881fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_model(model_name, y_true, y_pred, test_gen):\n",
    "    labels = list(test_gen.class_indices.keys())\n",
    "    accuracy = accuracy_score(y_true, y_pred)\n",
    "    precision = precision_score(y_true, y_pred, average='weighted')\n",
    "    recall = recall_score(y_true, y_pred, average='weighted')\n",
    "    f1 = f1_score(y_true, y_pred, average='weighted')\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    \n",
    "    print(\"Model:\", model_name)\n",
    "    print(\"Accuracy:\", accuracy)\n",
    "    print(\"Precision:\", precision)\n",
    "    print(\"Recall:\", recall)\n",
    "    print(\"F1-score:\", f1)\n",
    "    \n",
    "    print(\"Classification Report:\")\n",
    "    print(classification_report(y_true, y_pred, target_names=labels))\n",
    "    \n",
    "    # Plot confusion matrix\n",
    "    plt.figure(figsize=(6, 5))\n",
    "    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=labels, yticklabels=labels)\n",
    "    plt.xlabel(\"Predicted Label\")\n",
    "    plt.ylabel(\"True Label\")\n",
    "    plt.title(\"Confusion Matrix\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f2cf08b2-7627-451e-88d5-a023e089f52c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1519 images belonging to 2 classes.\n",
      "Found 217 images belonging to 2 classes.\n",
      "Found 436 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "# Directories for datasets\n",
    "train_dir = 'dataset/train'\n",
    "val_dir = 'dataset/validation'\n",
    "test_dir = 'dataset/test'\n",
    "\n",
    "# Get input shape and number of classes\n",
    "input_shape = (240, 240, 3)\n",
    "\n",
    "# Create data generators\n",
    "train_gen, val_gen, test_gen = create_data_generators(train_dir, val_dir, test_dir)\n",
    "num_classes = len(train_gen.class_indices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a8e05758-dcf9-4a89-8081-7a3243a53904",
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_resnet8_model(input_shape, num_classes):\n",
    "    base_model = ResNet50(weights='imagenet', include_top=False, input_shape=input_shape)\n",
    "    base_model.trainable = False  # Freeze the base model\n",
    "    \n",
    "    model = Sequential([\n",
    "        base_model,\n",
    "        GlobalAveragePooling2D(),\n",
    "        Dense(128, activation='relu'),\n",
    "        Dropout(0.5),\n",
    "        Dense(num_classes, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    model.compile(optimizer='adam',\n",
    "                  loss='categorical_crossentropy',\n",
    "                  metrics=['accuracy'])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "531f584b-0ad9-43d6-8e0f-f957fe315799",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m66s\u001b[0m 1s/step - accuracy: 0.5741 - loss: 0.7052 - val_accuracy: 0.6682 - val_loss: 0.6534\n",
      "Epoch 2/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6005 - loss: 0.6583 - val_accuracy: 0.6728 - val_loss: 0.6420\n",
      "Epoch 3/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m64s\u001b[0m 1s/step - accuracy: 0.6434 - loss: 0.6451 - val_accuracy: 0.6359 - val_loss: 0.6325\n",
      "Epoch 4/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m64s\u001b[0m 1s/step - accuracy: 0.6633 - loss: 0.6312 - val_accuracy: 0.6728 - val_loss: 0.6233\n",
      "Epoch 5/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m62s\u001b[0m 1s/step - accuracy: 0.6447 - loss: 0.6406 - val_accuracy: 0.6774 - val_loss: 0.6189\n",
      "Epoch 6/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6579 - loss: 0.6308 - val_accuracy: 0.6912 - val_loss: 0.6317\n",
      "Epoch 7/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6774 - loss: 0.6127 - val_accuracy: 0.6728 - val_loss: 0.6093\n",
      "Epoch 8/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6757 - loss: 0.6048 - val_accuracy: 0.6774 - val_loss: 0.6216\n",
      "Epoch 9/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6963 - loss: 0.6036 - val_accuracy: 0.6728 - val_loss: 0.5943\n",
      "Epoch 10/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6624 - loss: 0.6039 - val_accuracy: 0.6728 - val_loss: 0.5959\n",
      "Epoch 11/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6792 - loss: 0.5955 - val_accuracy: 0.6774 - val_loss: 0.5984\n",
      "Epoch 12/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m66s\u001b[0m 1s/step - accuracy: 0.7097 - loss: 0.5726 - val_accuracy: 0.6866 - val_loss: 0.6108\n",
      "Epoch 13/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m66s\u001b[0m 1s/step - accuracy: 0.6825 - loss: 0.5912 - val_accuracy: 0.6866 - val_loss: 0.5966\n",
      "Epoch 14/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m62s\u001b[0m 1s/step - accuracy: 0.7069 - loss: 0.5731 - val_accuracy: 0.6912 - val_loss: 0.5932\n",
      "Epoch 15/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.7011 - loss: 0.5626 - val_accuracy: 0.6774 - val_loss: 0.5810\n",
      "Epoch 16/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6975 - loss: 0.5713 - val_accuracy: 0.6959 - val_loss: 0.6191\n",
      "Epoch 17/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m60s\u001b[0m 1s/step - accuracy: 0.6891 - loss: 0.5719 - val_accuracy: 0.6774 - val_loss: 0.5903\n",
      "Epoch 18/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.7017 - loss: 0.5721 - val_accuracy: 0.6820 - val_loss: 0.5796\n",
      "Epoch 19/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6910 - loss: 0.5978 - val_accuracy: 0.6728 - val_loss: 0.5802\n",
      "Epoch 20/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6709 - loss: 0.5817 - val_accuracy: 0.6866 - val_loss: 0.5981\n",
      "Epoch 21/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6885 - loss: 0.5967 - val_accuracy: 0.6866 - val_loss: 0.6078\n",
      "Epoch 22/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6391 - loss: 0.5930 - val_accuracy: 0.6774 - val_loss: 0.5986\n",
      "Epoch 23/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6606 - loss: 0.5705 - val_accuracy: 0.6866 - val_loss: 0.5951\n",
      "Epoch 24/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6516 - loss: 0.5827 - val_accuracy: 0.6774 - val_loss: 0.6068\n",
      "Epoch 25/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6620 - loss: 0.5853 - val_accuracy: 0.6866 - val_loss: 0.5816\n",
      "Epoch 26/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6712 - loss: 0.5869 - val_accuracy: 0.6636 - val_loss: 0.6002\n",
      "Epoch 27/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6639 - loss: 0.6115 - val_accuracy: 0.6866 - val_loss: 0.5861\n",
      "Epoch 28/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6660 - loss: 0.5956 - val_accuracy: 0.6774 - val_loss: 0.6071\n",
      "Epoch 29/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6635 - loss: 0.5956 - val_accuracy: 0.6866 - val_loss: 0.5754\n",
      "Epoch 30/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6868 - loss: 0.5743 - val_accuracy: 0.6866 - val_loss: 0.5682\n",
      "Epoch 31/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6769 - loss: 0.5606 - val_accuracy: 0.6544 - val_loss: 0.6053\n",
      "Epoch 32/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6835 - loss: 0.5768 - val_accuracy: 0.6820 - val_loss: 0.5718\n",
      "Epoch 33/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m60s\u001b[0m 1s/step - accuracy: 0.6505 - loss: 0.5727 - val_accuracy: 0.6820 - val_loss: 0.5631\n",
      "Epoch 34/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6716 - loss: 0.5709 - val_accuracy: 0.6912 - val_loss: 0.5746\n",
      "Epoch 35/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6968 - loss: 0.5547 - val_accuracy: 0.6912 - val_loss: 0.5764\n",
      "Epoch 36/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6643 - loss: 0.5864 - val_accuracy: 0.6912 - val_loss: 0.5757\n",
      "Epoch 37/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m63s\u001b[0m 1s/step - accuracy: 0.6788 - loss: 0.5623 - val_accuracy: 0.6636 - val_loss: 0.5848\n",
      "Epoch 38/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m73s\u001b[0m 2s/step - accuracy: 0.6556 - loss: 0.5709 - val_accuracy: 0.6866 - val_loss: 0.5563\n",
      "Epoch 39/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 2s/step - accuracy: 0.6754 - loss: 0.5328 - val_accuracy: 0.6866 - val_loss: 0.5531\n",
      "Epoch 40/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 2s/step - accuracy: 0.6973 - loss: 0.5601 - val_accuracy: 0.6728 - val_loss: 0.5769\n",
      "Epoch 41/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 2s/step - accuracy: 0.6753 - loss: 0.5601 - val_accuracy: 0.6590 - val_loss: 0.5745\n",
      "Epoch 42/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m82s\u001b[0m 2s/step - accuracy: 0.6890 - loss: 0.5642 - val_accuracy: 0.6820 - val_loss: 0.5850\n",
      "Epoch 43/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 2s/step - accuracy: 0.6700 - loss: 0.5888 - val_accuracy: 0.6728 - val_loss: 0.5877\n",
      "Epoch 44/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 2s/step - accuracy: 0.6974 - loss: 0.5591 - val_accuracy: 0.6866 - val_loss: 0.5576\n",
      "Epoch 45/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 2s/step - accuracy: 0.6692 - loss: 0.5656 - val_accuracy: 0.6636 - val_loss: 0.5676\n",
      "Epoch 46/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 2s/step - accuracy: 0.6762 - loss: 0.5628 - val_accuracy: 0.6912 - val_loss: 0.5575\n",
      "Epoch 47/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m74s\u001b[0m 2s/step - accuracy: 0.6824 - loss: 0.5571 - val_accuracy: 0.6590 - val_loss: 0.5712\n",
      "Epoch 48/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m70s\u001b[0m 1s/step - accuracy: 0.6795 - loss: 0.5653 - val_accuracy: 0.6313 - val_loss: 0.6325\n",
      "Epoch 49/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m66s\u001b[0m 1s/step - accuracy: 0.6568 - loss: 0.5862 - val_accuracy: 0.6452 - val_loss: 0.6128\n",
      "Epoch 50/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m61s\u001b[0m 1s/step - accuracy: 0.6915 - loss: 0.5617 - val_accuracy: 0.6590 - val_loss: 0.5712\n"
     ]
    }
   ],
   "source": [
    "model = build_resnet8_model(input_shape, num_classes)\n",
    "\n",
    "history = model_resnet8.fit(\n",
    "    train_gen,\n",
    "    epochs=50,\n",
    "    validation_data=val_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e2cef2d1-332e-40b5-a942-ebc49c75210b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plot training & validation accuracy values\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Accuracy curve\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Train Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Model Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "# Loss curve\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Train Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Model Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "77c727b9-85cb-4efd-abcc-87f69cbb689c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 999ms/step\n",
      "Model: ResNet50\n",
      "Accuracy: 0.7270642201834863\n",
      "Precision: 0.7784018981238282\n",
      "Recall: 0.7270642201834863\n",
      "F1-score: 0.7146635644977923\n",
      "Classification Report:\n",
      "                      precision    recall  f1-score   support\n",
      "\n",
      "diabetic_retinopathy       0.90      0.52      0.66       220\n",
      "              normal       0.66      0.94      0.77       216\n",
      "\n",
      "            accuracy                           0.73       436\n",
      "           macro avg       0.78      0.73      0.72       436\n",
      "        weighted avg       0.78      0.73      0.71       436\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = model_resnet8.predict(test_gen)\n",
    "y_true = test_gen.classes\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "evaluate_model('ResNet50', y_true, y_pred_classes, test_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "71bedc3a-8233-48f5-b8e3-e755c2549567",
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_efficientnetb0_model(input_shape, num_classes):\n",
    "    base_model = EfficientNetB0(weights='imagenet', include_top=False, input_shape=input_shape)  \n",
    "    base_model.trainable = False  # Freeze the base model\n",
    "    \n",
    "    model = Sequential([\n",
    "        base_model,\n",
    "        GlobalAveragePooling2D(),\n",
    "        Dense(128, activation='relu'),\n",
    "        Dropout(0.5),\n",
    "        Dense(num_classes, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    model.compile(optimizer='adam',\n",
    "                  loss='categorical_crossentropy',\n",
    "                  metrics=['accuracy'])\n",
    "    \n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e26d7803-5821-40da-bbb8-188d92023df3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m50s\u001b[0m 899ms/step - accuracy: 0.4728 - loss: 0.7641 - val_accuracy: 0.4931 - val_loss: 0.6943\n",
      "Epoch 2/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 905ms/step - accuracy: 0.4696 - loss: 0.7030 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 3/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 849ms/step - accuracy: 0.5016 - loss: 0.6926 - val_accuracy: 0.4931 - val_loss: 0.6932\n",
      "Epoch 4/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 803ms/step - accuracy: 0.5063 - loss: 0.6953 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 5/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 799ms/step - accuracy: 0.5037 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6930\n",
      "Epoch 6/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m38s\u001b[0m 797ms/step - accuracy: 0.5112 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 7/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 801ms/step - accuracy: 0.5017 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 8/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m38s\u001b[0m 796ms/step - accuracy: 0.4870 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 9/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m38s\u001b[0m 795ms/step - accuracy: 0.5129 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 10/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 804ms/step - accuracy: 0.5017 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 11/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 799ms/step - accuracy: 0.5112 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 12/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 807ms/step - accuracy: 0.5001 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 13/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 802ms/step - accuracy: 0.4903 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 14/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 802ms/step - accuracy: 0.5065 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 15/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 937ms/step - accuracy: 0.5024 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 16/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 936ms/step - accuracy: 0.4757 - loss: 0.6937 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 17/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 827ms/step - accuracy: 0.4956 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 18/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m38s\u001b[0m 799ms/step - accuracy: 0.5209 - loss: 0.6927 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 19/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.5101 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 20/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 803ms/step - accuracy: 0.4892 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 21/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 806ms/step - accuracy: 0.5197 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 22/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 812ms/step - accuracy: 0.5010 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 23/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 806ms/step - accuracy: 0.5073 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 24/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 809ms/step - accuracy: 0.5319 - loss: 0.6926 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 25/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 804ms/step - accuracy: 0.5056 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 26/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 800ms/step - accuracy: 0.5075 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 27/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 807ms/step - accuracy: 0.4959 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 28/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 806ms/step - accuracy: 0.5200 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 29/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 811ms/step - accuracy: 0.4855 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 30/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 801ms/step - accuracy: 0.5129 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 31/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 810ms/step - accuracy: 0.5131 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 32/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 809ms/step - accuracy: 0.4938 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 33/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.4992 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 34/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 806ms/step - accuracy: 0.5250 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 35/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 802ms/step - accuracy: 0.5183 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 36/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 856ms/step - accuracy: 0.5068 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 37/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.5071 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 38/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m47s\u001b[0m 977ms/step - accuracy: 0.5124 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 39/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 967ms/step - accuracy: 0.4904 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 40/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.5049 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 41/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m50s\u001b[0m 1s/step - accuracy: 0.5125 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 42/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m47s\u001b[0m 968ms/step - accuracy: 0.4885 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 43/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 951ms/step - accuracy: 0.5083 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 44/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m50s\u001b[0m 1s/step - accuracy: 0.5058 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 45/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m50s\u001b[0m 1s/step - accuracy: 0.4974 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 46/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.5189 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 47/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.4895 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 48/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.5237 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 49/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.5059 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 50/50\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m53s\u001b[0m 1s/step - accuracy: 0.5186 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n"
     ]
    }
   ],
   "source": [
    "efficientnetb0_model = build_efficientnetb0_model(input_shape, num_classes)\n",
    "\n",
    "history = efficientnetb0_model.fit(\n",
    "    train_gen,\n",
    "    epochs=50,\n",
    "    validation_data=val_gen\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ccc52e61-fb01-4ae8-8c0a-29db16abff77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt \n",
    "\n",
    "# Plot training & validation accuracy values\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Accuracy curve\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Train Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Model Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "# Loss curve\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Train Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Model Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "0c59b75e-010e-4279-ac60-2d35e7096ae8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 485ms/step\n",
      "Model: EfficientNetB0\n",
      "Accuracy: 0.5045871559633027\n",
      "Precision: 0.2546081979631344\n",
      "Recall: 0.5045871559633027\n",
      "F1-score: 0.33844260460953235\n",
      "Classification Report:\n",
      "                      precision    recall  f1-score   support\n",
      "\n",
      "diabetic_retinopathy       0.50      1.00      0.67       220\n",
      "              normal       0.00      0.00      0.00       216\n",
      "\n",
      "            accuracy                           0.50       436\n",
      "           macro avg       0.25      0.50      0.34       436\n",
      "        weighted avg       0.25      0.50      0.34       436\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Downloads\\ANACONDA\\envs\\first1\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1565: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "D:\\Downloads\\ANACONDA\\envs\\first1\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1565: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "D:\\Downloads\\ANACONDA\\envs\\first1\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1565: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "D:\\Downloads\\ANACONDA\\envs\\first1\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1565: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = efficientnetb0_model.predict(test_gen)\n",
    "y_true = test_gen.classes\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "\n",
    "evaluate_model('EfficientNetB0', y_true, y_pred_classes, test_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d22295eb-6fff-401b-a2e3-51610dccdc62",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique predicted classes: [0]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "unique_preds = np.unique(y_pred_classes)\n",
    "print(\"Unique predicted classes:\", unique_preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "1eedde67-a4da-4349-aeed-9199afb1d26f",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    rotation_range=45,\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    zoom_range=0.2,\n",
    "    horizontal_flip=True,\n",
    "    vertical_flip=True,\n",
    "    brightness_range=(0.5, 1.5)  # Increase variation\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "dd4cc40f-76c1-42c0-bf87-fd9d13cd9443",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m57s\u001b[0m 1s/step - accuracy: 0.4821 - loss: 0.6937 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 2/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 951ms/step - accuracy: 0.4770 - loss: 0.6936 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 3/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 891ms/step - accuracy: 0.5081 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 4/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 897ms/step - accuracy: 0.5150 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 5/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 930ms/step - accuracy: 0.5060 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 6/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 941ms/step - accuracy: 0.5024 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 7/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 916ms/step - accuracy: 0.4990 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 8/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 904ms/step - accuracy: 0.5008 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 9/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 937ms/step - accuracy: 0.4974 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 10/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 936ms/step - accuracy: 0.4907 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 11/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 932ms/step - accuracy: 0.5047 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 12/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 883ms/step - accuracy: 0.5167 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 13/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 819ms/step - accuracy: 0.4948 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 14/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 800ms/step - accuracy: 0.5086 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 15/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 830ms/step - accuracy: 0.4969 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 16/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 915ms/step - accuracy: 0.5122 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 17/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 864ms/step - accuracy: 0.4985 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 18/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 822ms/step - accuracy: 0.5175 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 19/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 818ms/step - accuracy: 0.4993 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 20/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 826ms/step - accuracy: 0.5092 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 21/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 837ms/step - accuracy: 0.5140 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 22/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 858ms/step - accuracy: 0.5347 - loss: 0.6924 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 23/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m49s\u001b[0m 1s/step - accuracy: 0.4964 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 24/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 854ms/step - accuracy: 0.5103 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 25/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 857ms/step - accuracy: 0.5043 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 26/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 813ms/step - accuracy: 0.4961 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 27/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 868ms/step - accuracy: 0.4967 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 28/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 830ms/step - accuracy: 0.4984 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 29/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 842ms/step - accuracy: 0.5008 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 30/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 809ms/step - accuracy: 0.4885 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 31/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 813ms/step - accuracy: 0.5220 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 32/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 820ms/step - accuracy: 0.4935 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 33/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 805ms/step - accuracy: 0.4976 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 34/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 849ms/step - accuracy: 0.5025 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 35/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 863ms/step - accuracy: 0.5060 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 36/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 835ms/step - accuracy: 0.5141 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 37/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 895ms/step - accuracy: 0.5170 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 38/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 857ms/step - accuracy: 0.5189 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 39/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.5078 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 40/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 811ms/step - accuracy: 0.5102 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 41/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 811ms/step - accuracy: 0.4895 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 42/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 820ms/step - accuracy: 0.5196 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 43/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 815ms/step - accuracy: 0.4862 - loss: 0.6936 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 44/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 823ms/step - accuracy: 0.5076 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 45/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 881ms/step - accuracy: 0.4891 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 46/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 851ms/step - accuracy: 0.5145 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 47/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 877ms/step - accuracy: 0.5038 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 48/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 869ms/step - accuracy: 0.4798 - loss: 0.6936 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 49/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 864ms/step - accuracy: 0.5245 - loss: 0.6927 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 50/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 857ms/step - accuracy: 0.5011 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 51/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 860ms/step - accuracy: 0.5155 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 52/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 861ms/step - accuracy: 0.4915 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 53/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 863ms/step - accuracy: 0.5132 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 54/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 878ms/step - accuracy: 0.4931 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 55/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 826ms/step - accuracy: 0.5181 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 56/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 815ms/step - accuracy: 0.5147 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 57/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 807ms/step - accuracy: 0.4979 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 58/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 809ms/step - accuracy: 0.5210 - loss: 0.6927 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 59/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 807ms/step - accuracy: 0.4828 - loss: 0.6937 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 60/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.5230 - loss: 0.6927 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 61/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 810ms/step - accuracy: 0.5051 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 62/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 810ms/step - accuracy: 0.5159 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 63/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 816ms/step - accuracy: 0.5077 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 64/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 810ms/step - accuracy: 0.4996 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 65/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 812ms/step - accuracy: 0.4820 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 66/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 812ms/step - accuracy: 0.5058 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 67/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 812ms/step - accuracy: 0.5166 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 68/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 829ms/step - accuracy: 0.5113 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 69/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 814ms/step - accuracy: 0.5115 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 70/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 811ms/step - accuracy: 0.5028 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 71/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 813ms/step - accuracy: 0.4792 - loss: 0.6937 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 72/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 818ms/step - accuracy: 0.5055 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 73/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.4979 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 74/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 817ms/step - accuracy: 0.4936 - loss: 0.6935 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 75/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.5248 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 76/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 826ms/step - accuracy: 0.5137 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 77/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 812ms/step - accuracy: 0.5051 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 78/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 814ms/step - accuracy: 0.4892 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 79/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 833ms/step - accuracy: 0.4958 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 80/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 854ms/step - accuracy: 0.5180 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 81/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 815ms/step - accuracy: 0.5033 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 82/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 821ms/step - accuracy: 0.4980 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 83/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 818ms/step - accuracy: 0.5055 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 84/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 811ms/step - accuracy: 0.5119 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 85/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 809ms/step - accuracy: 0.4821 - loss: 0.6938 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 86/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 808ms/step - accuracy: 0.5178 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 87/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 839ms/step - accuracy: 0.5090 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 88/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 848ms/step - accuracy: 0.5162 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 89/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 813ms/step - accuracy: 0.4990 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 90/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 861ms/step - accuracy: 0.5041 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 91/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m42s\u001b[0m 876ms/step - accuracy: 0.5143 - loss: 0.6930 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 92/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 962ms/step - accuracy: 0.5055 - loss: 0.6931 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 93/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 939ms/step - accuracy: 0.4980 - loss: 0.6933 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 94/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 918ms/step - accuracy: 0.5129 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 95/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 930ms/step - accuracy: 0.5193 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 96/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 937ms/step - accuracy: 0.5125 - loss: 0.6929 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 97/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 914ms/step - accuracy: 0.5016 - loss: 0.6932 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 98/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 920ms/step - accuracy: 0.4947 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 99/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 945ms/step - accuracy: 0.5176 - loss: 0.6928 - val_accuracy: 0.5069 - val_loss: 0.6931\n",
      "Epoch 100/100\n",
      "\u001b[1m48/48\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 924ms/step - accuracy: 0.4996 - loss: 0.6934 - val_accuracy: 0.5069 - val_loss: 0.6931\n"
     ]
    }
   ],
   "source": [
    "history = efficientnetb0_model.fit(train_gen, epochs=100, validation_data=val_gen)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "4d4a9331-a841-4766-b9a5-fa8e832c1978",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt \n",
    "\n",
    "# Plot training & validation accuracy values\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Accuracy curve\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Train Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Model Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "# Loss curve\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Train Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Model Loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "a1d5b78e-1458-4039-b133-a46ca744bb25",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 587ms/step\n",
      "Model: EfficientNetB0\n",
      "Accuracy: 0.5045871559633027\n",
      "Precision: 0.7500210419998317\n",
      "Recall: 0.5045871559633027\n",
      "F1-score: 0.33844260460953235\n",
      "Classification Report:\n",
      "                      precision    recall  f1-score   support\n",
      "\n",
      "diabetic_retinopathy       0.50      1.00      0.67       220\n",
      "              normal       1.00      0.00      0.00       216\n",
      "\n",
      "            accuracy                           0.50       436\n",
      "           macro avg       0.75      0.50      0.34       436\n",
      "        weighted avg       0.75      0.50      0.34       436\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = efficientnetb0_model.predict(test_gen)\n",
    "y_true = test_gen.classes\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "\n",
    "evaluate_model('EfficientNetB0', y_true, y_pred_classes, test_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "df8fb657-4730-4e54-a073-dfc5a6b49436",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}