tl.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix, classification_report, roc_curve, auc\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define image dimensions and batch size\n",
    "img_width, img_height = 200, 200\n",
    "batch_size = 32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract class labels from folder names in the training directory\n",
    "train_dir = 'training'\n",
    "class_names = sorted(os.listdir(train_dir))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create data generators with increased data augmentation for the training set\n",
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    shear_range=0.2,\n",
    "    zoom_range=0.2,\n",
    "    rotation_range=40,  # Increase rotation range\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    brightness_range=[0.5, 1.5],  # Increase brightness range\n",
    "    horizontal_flip=True,\n",
    "    validation_split=0.2\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 4571 images belonging to 4 classes.\n",
      "Found 1141 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "# Data generators for training and validation\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "    train_dir,\n",
    "    target_size=(img_width, img_height),\n",
    "    batch_size=batch_size,\n",
    "    class_mode='categorical',\n",
    "    subset='training',\n",
    "    \n",
    ")\n",
    "\n",
    "validation_generator = train_datagen.flow_from_directory(\n",
    "    train_dir,\n",
    "    target_size=(img_width, img_height),\n",
    "    batch_size=batch_size,\n",
    "    class_mode='categorical',\n",
    "    subset='validation'\n",
    ")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:`input_shape` is undefined or non-square, or `rows` is not in [96, 128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.\n"
     ]
    }
   ],
   "source": [
    "# Load MobileNetV2 model\n",
    "base_model = tf.keras.applications.MobileNetV2(input_shape=(img_width, img_height, 3),\n",
    "                                               include_top=False,\n",
    "                                               weights='imagenet')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fine-tune MobileNetV2 by unfreezing some layers\n",
    "base_model.trainable = True\n",
    "\n",
    "# Fine-tune from this layer onwards\n",
    "fine_tune_at = 100\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Freeze all layers before the fine-tune layer\n",
    "for layer in base_model.layers[:fine_tune_at]:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build the model\n",
    "model = models.Sequential([\n",
    "    base_model,\n",
    "    layers.GlobalAveragePooling2D(),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(512, activation='relu'),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(len(class_names), activation='softmax')\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compile the model with a lower learning rate\n",
    "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001),\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Early stopping\n",
    "early_stopping = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/40\n",
      "142/142 [==============================] - 113s 765ms/step - loss: 0.7277 - accuracy: 0.7118 - val_loss: 1.6872 - val_accuracy: 0.5196\n",
      "Epoch 2/40\n",
      "142/142 [==============================] - 108s 763ms/step - loss: 0.3409 - accuracy: 0.8735 - val_loss: 1.0871 - val_accuracy: 0.6420\n",
      "Epoch 3/40\n",
      "142/142 [==============================] - 104s 734ms/step - loss: 0.2588 - accuracy: 0.9057 - val_loss: 0.7055 - val_accuracy: 0.7473\n",
      "Epoch 4/40\n",
      "142/142 [==============================] - 103s 723ms/step - loss: 0.2007 - accuracy: 0.9266 - val_loss: 0.5773 - val_accuracy: 0.8188\n",
      "Epoch 5/40\n",
      "142/142 [==============================] - 103s 727ms/step - loss: 0.1698 - accuracy: 0.9370 - val_loss: 0.3639 - val_accuracy: 0.8857\n",
      "Epoch 6/40\n",
      "142/142 [==============================] - 109s 764ms/step - loss: 0.1484 - accuracy: 0.9471 - val_loss: 0.4680 - val_accuracy: 0.8679\n",
      "Epoch 7/40\n",
      "142/142 [==============================] - 108s 760ms/step - loss: 0.1443 - accuracy: 0.9493 - val_loss: 0.4587 - val_accuracy: 0.8705\n",
      "Epoch 8/40\n",
      "142/142 [==============================] - 109s 770ms/step - loss: 0.1310 - accuracy: 0.9520 - val_loss: 0.3543 - val_accuracy: 0.8938\n",
      "Epoch 9/40\n",
      "142/142 [==============================] - 106s 745ms/step - loss: 0.1000 - accuracy: 0.9672 - val_loss: 0.5340 - val_accuracy: 0.8571\n",
      "Epoch 10/40\n",
      "142/142 [==============================] - 103s 722ms/step - loss: 0.0810 - accuracy: 0.9722 - val_loss: 0.4208 - val_accuracy: 0.8857\n",
      "Epoch 11/40\n",
      "142/142 [==============================] - 105s 739ms/step - loss: 0.0973 - accuracy: 0.9692 - val_loss: 0.5055 - val_accuracy: 0.8750\n",
      "Epoch 12/40\n",
      "142/142 [==============================] - 103s 726ms/step - loss: 0.0854 - accuracy: 0.9709 - val_loss: 0.3904 - val_accuracy: 0.8955\n",
      "Epoch 13/40\n",
      "142/142 [==============================] - 103s 727ms/step - loss: 0.0752 - accuracy: 0.9738 - val_loss: 0.3966 - val_accuracy: 0.8929\n",
      "Epoch 14/40\n",
      "142/142 [==============================] - 107s 755ms/step - loss: 0.0684 - accuracy: 0.9788 - val_loss: 0.3678 - val_accuracy: 0.8955\n",
      "Epoch 15/40\n",
      "142/142 [==============================] - 105s 741ms/step - loss: 0.0662 - accuracy: 0.9775 - val_loss: 0.3432 - val_accuracy: 0.9089\n",
      "Epoch 16/40\n",
      "142/142 [==============================] - 108s 759ms/step - loss: 0.0585 - accuracy: 0.9800 - val_loss: 0.3224 - val_accuracy: 0.9116\n",
      "Epoch 17/40\n",
      "142/142 [==============================] - 101s 710ms/step - loss: 0.0679 - accuracy: 0.9769 - val_loss: 0.4450 - val_accuracy: 0.8821\n",
      "Epoch 18/40\n",
      "142/142 [==============================] - 102s 718ms/step - loss: 0.0633 - accuracy: 0.9771 - val_loss: 0.2942 - val_accuracy: 0.9232\n",
      "Epoch 19/40\n",
      "142/142 [==============================] - 101s 711ms/step - loss: 0.0514 - accuracy: 0.9846 - val_loss: 0.3521 - val_accuracy: 0.9214\n",
      "Epoch 20/40\n",
      "142/142 [==============================] - 101s 710ms/step - loss: 0.0462 - accuracy: 0.9848 - val_loss: 0.2645 - val_accuracy: 0.9312\n",
      "Epoch 21/40\n",
      "142/142 [==============================] - 101s 711ms/step - loss: 0.0485 - accuracy: 0.9835 - val_loss: 0.2908 - val_accuracy: 0.9286\n",
      "Epoch 22/40\n",
      "142/142 [==============================] - 101s 709ms/step - loss: 0.0426 - accuracy: 0.9870 - val_loss: 0.2750 - val_accuracy: 0.9268\n",
      "Epoch 23/40\n",
      "142/142 [==============================] - 102s 715ms/step - loss: 0.0474 - accuracy: 0.9835 - val_loss: 0.2659 - val_accuracy: 0.9429\n",
      "Epoch 24/40\n",
      "142/142 [==============================] - 104s 731ms/step - loss: 0.0574 - accuracy: 0.9811 - val_loss: 0.2698 - val_accuracy: 0.9330\n",
      "Epoch 25/40\n",
      "142/142 [==============================] - 105s 738ms/step - loss: 0.0467 - accuracy: 0.9857 - val_loss: 0.2817 - val_accuracy: 0.9312\n",
      "Epoch 26/40\n",
      "142/142 [==============================] - 105s 740ms/step - loss: 0.0342 - accuracy: 0.9888 - val_loss: 0.2695 - val_accuracy: 0.9241\n",
      "Epoch 27/40\n",
      "142/142 [==============================] - 103s 727ms/step - loss: 0.0447 - accuracy: 0.9868 - val_loss: 0.2579 - val_accuracy: 0.9196\n",
      "Epoch 28/40\n",
      "142/142 [==============================] - 113s 796ms/step - loss: 0.0405 - accuracy: 0.9861 - val_loss: 0.2626 - val_accuracy: 0.9384\n",
      "Epoch 29/40\n",
      "142/142 [==============================] - 125s 878ms/step - loss: 0.0301 - accuracy: 0.9894 - val_loss: 0.3808 - val_accuracy: 0.9054\n",
      "Epoch 30/40\n",
      "142/142 [==============================] - 103s 727ms/step - loss: 0.0476 - accuracy: 0.9850 - val_loss: 0.3231 - val_accuracy: 0.9295\n",
      "Epoch 31/40\n",
      "142/142 [==============================] - 104s 731ms/step - loss: 0.0440 - accuracy: 0.9879 - val_loss: 0.3642 - val_accuracy: 0.9179\n",
      "Epoch 32/40\n",
      "142/142 [==============================] - 107s 753ms/step - loss: 0.0369 - accuracy: 0.9868 - val_loss: 0.3049 - val_accuracy: 0.9304\n",
      "Epoch 33/40\n",
      "142/142 [==============================] - 104s 731ms/step - loss: 0.0289 - accuracy: 0.9896 - val_loss: 0.3423 - val_accuracy: 0.9250\n",
      "Epoch 34/40\n",
      "142/142 [==============================] - 105s 737ms/step - loss: 0.0283 - accuracy: 0.9914 - val_loss: 0.3292 - val_accuracy: 0.9196\n",
      "Epoch 35/40\n",
      "142/142 [==============================] - 105s 739ms/step - loss: 0.0272 - accuracy: 0.9907 - val_loss: 0.3432 - val_accuracy: 0.9223\n",
      "Epoch 36/40\n",
      "142/142 [==============================] - 111s 779ms/step - loss: 0.0302 - accuracy: 0.9899 - val_loss: 0.3436 - val_accuracy: 0.9339\n",
      "Epoch 37/40\n",
      "142/142 [==============================] - 107s 755ms/step - loss: 0.0284 - accuracy: 0.9907 - val_loss: 0.2681 - val_accuracy: 0.9384\n"
     ]
    }
   ],
   "source": [
    "# Train the model for more epochs\n",
    "epochs = 40\n",
    "history = model.fit(\n",
    "    train_generator,\n",
    "    steps_per_epoch=train_generator.samples // batch_size,\n",
    "    validation_data=validation_generator,\n",
    "    validation_steps=validation_generator.samples // batch_size,\n",
    "    epochs=epochs,\n",
    "    callbacks=[early_stopping]\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1311 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model on the test set\n",
    "test_generator = train_datagen.flow_from_directory(\n",
    "    'testing',\n",
    "    target_size=(img_width, img_height),\n",
    "    batch_size=batch_size,\n",
    "    class_mode='categorical',\n",
    "    shuffle=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 26s 612ms/step\n"
     ]
    }
   ],
   "source": [
    "# Get true and predicted labels\n",
    "y_true = test_generator.classes\n",
    "y_prob = model.predict(test_generator)\n",
    "y_pred = np.argmax(y_prob, axis=-1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "      glioma       0.96      0.92      0.94       300\n",
      "  meningioma       0.92      0.90      0.91       306\n",
      "     notumor       0.96      0.99      0.97       405\n",
      "   pituitary       0.96      0.99      0.97       300\n",
      "\n",
      "    accuracy                           0.95      1311\n",
      "   macro avg       0.95      0.95      0.95      1311\n",
      "weighted avg       0.95      0.95      0.95      1311\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Compute confusion matrix\n",
    "conf_matrix = confusion_matrix(y_true, y_pred)\n",
    "\n",
    "# Plot confusion matrix with folder names as labels\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=class_names, yticklabels=class_names)\n",
    "plt.xlabel('Predicted Label')\n",
    "plt.ylabel('True Label')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n",
    "\n",
    "# Classification Report with folder names as labels\n",
    "print(\"Classification Report:\")\n",
    "print(classification_report(y_true, y_pred, target_names=class_names))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kalpa\\AppData\\Local\\Temp\\ipykernel_22856\\3140352687.py:14: UserWarning: color is redundantly defined by the 'color' keyword argument and the fmt string \"k--\" (-> color='k'). The keyword argument will take precedence.\n",
      "  plt.plot([0, 1], [0, 1], 'k--', color='gray', label='Random Guessing')\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Compute ROC curve and ROC area for each class\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "for i in range(len(class_names)):\n",
    "    fpr[i], tpr[i], _ = roc_curve(y_true == i, y_prob[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "# %%\n",
    "# Plot ROC curve for each class\n",
    "plt.figure(figsize=(10, 8))\n",
    "for i in range(len(class_names)):\n",
    "    plt.plot(fpr[i], tpr[i], label=f'{class_names[i]} (AUC = {roc_auc[i]:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', color='gray', label='Random Guessing')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kalpa\\AppData\\Roaming\\Python\\Python311\\site-packages\\keras\\src\\engine\\training.py:3079: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
      "  saving_api.save_model(\n"
     ]
    }
   ],
   "source": [
    "# Save the model\n",
    "model.save('mobilenetv2_tumor_detection_model.h5')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}