diff --git "a/3.RESNET50 Tumor Detection.ipynb" "b/3.RESNET50 Tumor Detection.ipynb" new file mode 100644--- /dev/null +++ "b/3.RESNET50 Tumor Detection.ipynb" @@ -0,0 +1 @@ +{"cells":[{"cell_type":"markdown","source":["# **Brain MRI Tumor Classification Using ResNet50 Transfer Learning**\n","\n","ResNet50 was evaluated using the same dataset and preprocessing pipeline."],"metadata":{"id":"kFRakAU-uEOm"}},{"cell_type":"markdown","source":["# **Architecture**\n","\n","ResNet50 (ImageNet pretrained) → GlobalAveragePooling2D → Dense(256) + BatchNorm + Dropout(0.4) → Dense(128) + BatchNorm + Dropout(0.3) → Dense(4, Softmax)"],"metadata":{"id":"BFzmHRHRwScS"}},{"cell_type":"markdown","source":["# **Final Result**\n","\n","Initial accuracy: **68.00%** , Fine-tuned accuracy: **81.00%** \n","\n","Fine-tuning produced a substantial **13 percentage**-point improvement, but the final performance remained below VGG16."],"metadata":{"id":"4MC87wtbyW3I"}},{"cell_type":"markdown","metadata":{"id":"qjQrXUwrIZdN"},"source":["**Libraries and datasets**"]},{"cell_type":"code","execution_count":1,"metadata":{"executionInfo":{"elapsed":5755,"status":"ok","timestamp":1787456167071,"user":{"displayName":"Abin","userId":"06944942458087513678"},"user_tz":-330},"id":"dVQpaeHGI1eK"},"outputs":[],"source":["import tensorflow as tf\n","from tensorflow.keras.models import Sequential\n","from tensorflow.keras.layers import Input, Flatten, Dropout, Dense, GlobalAveragePooling2D, BatchNormalization\n","from tensorflow.keras.optimizers import Adam\n","from tensorflow.keras.applications import ResNet50"]},{"cell_type":"code","source":["import os\n","import random"],"metadata":{"id":"cRXa2A8v7Lmy","executionInfo":{"status":"ok","timestamp":1787456167076,"user_tz":-330,"elapsed":14,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":2,"outputs":[]},{"cell_type":"code","source":["from PIL import Image, ImageEnhance\n","from tensorflow.keras.preprocessing.image import load_img\n","from sklearn.utils import shuffle"],"metadata":{"id":"ziTpmjh_7I25","executionInfo":{"status":"ok","timestamp":1787456167078,"user_tz":-330,"elapsed":11,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":3,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"qsuWozrTMP4R","executionInfo":{"status":"ok","timestamp":1787456167082,"user_tz":-330,"elapsed":3,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":3,"outputs":[]},{"cell_type":"code","source":["import numpy as np\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","from sklearn.metrics import classification_report, confusion_matrix, roc_curve, auc\n","from sklearn.preprocessing import label_binarize\n","from tensorflow.keras.models import load_model"],"metadata":{"id":"nxAAPGa757Ex","executionInfo":{"status":"ok","timestamp":1787456168187,"user_tz":-330,"elapsed":1102,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":4,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"yqVnKH849X3f","executionInfo":{"status":"ok","timestamp":1787456168195,"user_tz":-330,"elapsed":4,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":4,"outputs":[]},{"cell_type":"code","source":["print(\"GPU:\", tf.config.list_physical_devices('GPU'))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"t2yKNLVc55Z4","executionInfo":{"status":"ok","timestamp":1787456169836,"user_tz":-330,"elapsed":10,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"a5a14fb6-c810-4ed9-d0d0-ae27d5c506fb"},"execution_count":5,"outputs":[{"output_type":"stream","name":"stdout","text":["GPU: [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]\n"]}]},{"cell_type":"code","execution_count":6,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"J3yE0IytKG1c","executionInfo":{"status":"ok","timestamp":1787456198267,"user_tz":-330,"elapsed":28429,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"c0f9e0bd-52c5-4342-c1cb-f44dd97760ab"},"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}],"source":["from google.colab import drive\n","drive.mount('/content/drive')"]},{"cell_type":"code","source":["train_dir = '/content/drive/MyDrive/cancer-cnn/extracted data/Training'\n","test_dir = '/content/drive/MyDrive/cancer-cnn/extracted data/Testing'"],"metadata":{"id":"ZoFscuOz6Jv8","executionInfo":{"status":"ok","timestamp":1787456241840,"user_tz":-330,"elapsed":24,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":7,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"UrsAzHg66PfS","executionInfo":{"status":"ok","timestamp":1787456242737,"user_tz":-330,"elapsed":205,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":7,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"r2cZxSxzMgSd"},"source":["**preprocessing img**"]},{"cell_type":"code","source":["#img Augmentation fun\n","def augment_image(image):\n"," image = Image.fromarray(np.uint8(image))\n"," image = ImageEnhance.Brightness(image).enhance(random.uniform(0.8, 1.2))#brightness -random\n"," image = ImageEnhance.Contrast(image).enhance(random.uniform(0.8, 1.2)) #contrast -random\n"," image = np.array(image) / 255.0 #normalize pixel values to [0, 1]\n"," return image\n","\n","#load img & apply augmentation\n","def open_images(paths):\n"," images = []\n"," for path in paths:\n"," image = load_img(path, target_size=(IMAGE_SIZE, IMAGE_SIZE))\n"," image = augment_image(image)\n"," images.append(image)\n"," return np.array(images)\n","\n","#encoding labels (label names)\n","def encode_label(labels):\n"," unique_labels = os.listdir(train_dir) # Ensure unique labels are determined\n"," encoded = [unique_labels.index(label) for label in labels]\n"," return np.array(encoded)\n","\n","#data generator for batching\n","def datagen(paths, labels, batch_size=12, epochs=1):\n"," for _ in range(epochs):\n"," for i in range(0, len(paths), batch_size):\n"," batch_paths = paths[i:i + batch_size]\n"," batch_images = open_images(batch_paths) #open & augment img\n"," batch_labels = labels[i:i + batch_size]\n"," batch_labels = encode_label(batch_labels) # encode labels\n"," yield batch_images, batch_labels #yield the batch\n","\n"],"metadata":{"id":"fpQW8CvY9OZQ","executionInfo":{"status":"ok","timestamp":1787457256892,"user_tz":-330,"elapsed":25,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":15,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"NEyd1bjLKnZx","executionInfo":{"status":"ok","timestamp":1787457257772,"user_tz":-330,"elapsed":20,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":15,"outputs":[]},{"cell_type":"code","source":["from sklearn.model_selection import train_test_split\n","\n","train_paths, val_paths, train_labels, val_labels = train_test_split(\n"," train_paths,\n"," train_labels,\n"," test_size=0.15,\n"," random_state=42,\n"," stratify=train_labels\n",")"],"metadata":{"id":"Uxq6eYo_NjQK","executionInfo":{"status":"ok","timestamp":1787457258412,"user_tz":-330,"elapsed":30,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":16,"outputs":[]},{"cell_type":"code","source":["# FIXED CLASS ORDER\n","class_names = [\n"," \"pituitary\",\n"," \"notumor\",\n"," \"meningioma\",\n"," \"glioma\"\n","]\n","\n","print(\"Class mapping:\")\n","for i, name in enumerate(class_names):\n"," print(f\"{i} → {name}\")\n","\n","\n","# TRAIN DATA\n","train_paths = []\n","train_labels = []\n","\n","# Use our fixed class order, NOT os.listdir() order\n","for label in class_names:\n"," for image in os.listdir(os.path.join(train_dir, label)):\n"," train_paths.append(\n"," os.path.join(train_dir, label, image)\n"," )\n"," train_labels.append(label)\n","\n","train_paths, train_labels = shuffle(\n"," train_paths,\n"," train_labels,\n"," random_state=42\n",")\n","\n","\n","# CREATE VALIDATION SET\n","from sklearn.model_selection import train_test_split\n","\n","train_paths, val_paths, train_labels, val_labels = train_test_split(\n"," train_paths,\n"," train_labels,\n"," test_size=0.15,\n"," random_state=42,\n"," stratify=train_labels\n",")\n","\n","\n","# TEST DATA\n","test_paths = []\n","test_labels = []\n","\n","# Use the SAME fixed class order\n","for label in class_names:\n"," for image in os.listdir(os.path.join(test_dir, label)):\n"," test_paths.append(\n"," os.path.join(test_dir, label, image)\n"," )\n"," test_labels.append(label)\n","\n","test_paths, test_labels = shuffle(\n"," test_paths,\n"," test_labels,\n"," random_state=42\n",")\n","\n","\n","# LABEL ENCODING\n","def encode_label(labels):\n"," encoded = [\n"," class_names.index(label)\n"," for label in labels\n"," ]\n","\n"," return np.array(encoded)"],"metadata":{"id":"L4XDljU4NnUS","executionInfo":{"status":"ok","timestamp":1787457260307,"user_tz":-330,"elapsed":177,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"564c312a-6e9a-4e78-92a2-11889f89470d"},"execution_count":17,"outputs":[{"output_type":"stream","name":"stdout","text":["Class mapping:\n","0 → pituitary\n","1 → notumor\n","2 → meningioma\n","3 → glioma\n"]}]},{"cell_type":"code","source":["print(class_names)\n","print(encode_label(class_names))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GimB_ISWvl25","executionInfo":{"status":"ok","timestamp":1787457262988,"user_tz":-330,"elapsed":14,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"f8872e4d-5074-40f6-838c-cb685a7c192c"},"execution_count":18,"outputs":[{"output_type":"stream","name":"stdout","text":["['pituitary', 'notumor', 'meningioma', 'glioma']\n","[0 1 2 3]\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"mQ7OaRWYo89l"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# **Stage-1**\n","\n"," **Training of the model with first set of parameters**"],"metadata":{"id":"X3_kke9wKpkO"}},{"cell_type":"code","source":[],"metadata":{"id":"r5juY9eOs4Tm"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# MODEL 2 - ResNet50\n","\n","IMAGE_SIZE = 128\n","NUM_CLASSES = 4\n","\n","\n","# Load pretrained ResNet50\n","base_model = ResNet50(\n"," input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3),\n"," include_top=False,\n"," weights='imagenet'\n",")\n","\n","\n","# Freeze the complete ResNet50 backbone\n","for layer in base_model.layers:\n"," layer.trainable = False\n","\n","\n","# Classification head\n","model = Sequential()\n","\n","model.add(Input(shape=(IMAGE_SIZE, IMAGE_SIZE, 3)))\n","\n","model.add(base_model)\n","\n","model.add(GlobalAveragePooling2D())\n","\n","model.add(Dense(256, activation='relu'))\n","model.add(BatchNormalization())\n","model.add(Dropout(0.4))\n","\n","model.add(Dense(128, activation='relu'))\n","model.add(BatchNormalization())\n","model.add(Dropout(0.3))\n","\n","model.add(Dense(NUM_CLASSES, activation='softmax'))\n","\n","\n","# Compile\n","model.compile(\n"," optimizer=Adam(learning_rate=1e-4),\n"," loss='sparse_categorical_crossentropy',\n"," metrics=['sparse_categorical_accuracy']\n",")\n","\n","model.summary()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":520},"id":"qo5v50EwKnOv","executionInfo":{"status":"ok","timestamp":1787457309673,"user_tz":-330,"elapsed":4953,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"d48b50a8-1cf3-4d46-c85e-0e8f224b3b16"},"execution_count":19,"outputs":[{"output_type":"stream","name":"stdout","text":["Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5\n","\u001b[1m94765736/94765736\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n"]},{"output_type":"display_data","data":{"text/plain":["\u001b[1mModel: \"sequential\"\u001b[0m\n"],"text/html":["
Model: \"sequential\"\n","
\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ resnet50 (\u001b[38;5;33mFunctional\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m2048\u001b[0m) │ \u001b[38;5;34m23,587,712\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ global_average_pooling2d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n","│ (\u001b[38;5;33mGlobalAveragePooling2D\u001b[0m) │ │ │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m524,544\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ batch_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m1,024\u001b[0m │\n","│ (\u001b[38;5;33mBatchNormalization\u001b[0m) │ │ │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m32,896\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ batch_normalization_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m512\u001b[0m │\n","│ (\u001b[38;5;33mBatchNormalization\u001b[0m) │ │ │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_2 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m) │ \u001b[38;5;34m516\u001b[0m │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n"],"text/html":["
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n","┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n","┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n","│ resnet50 (Functional)           │ (None, 4, 4, 2048)     │    23,587,712 │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ global_average_pooling2d        │ (None, 2048)           │             0 │\n","│ (GlobalAveragePooling2D)        │                        │               │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense (Dense)                   │ (None, 256)            │       524,544 │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ batch_normalization             │ (None, 256)            │         1,024 │\n","│ (BatchNormalization)            │                        │               │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dropout (Dropout)               │ (None, 256)            │             0 │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_1 (Dense)                 │ (None, 128)            │        32,896 │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ batch_normalization_1           │ (None, 128)            │           512 │\n","│ (BatchNormalization)            │                        │               │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dropout_1 (Dropout)             │ (None, 128)            │             0 │\n","├─────────────────────────────────┼────────────────────────┼───────────────┤\n","│ dense_2 (Dense)                 │ (None, 4)              │           516 │\n","└─────────────────────────────────┴────────────────────────┴───────────────┘\n","
\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Total params: \u001b[0m\u001b[38;5;34m24,147,204\u001b[0m (92.11 MB)\n"],"text/html":["
 Total params: 24,147,204 (92.11 MB)\n","
\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m558,724\u001b[0m (2.13 MB)\n"],"text/html":["
 Trainable params: 558,724 (2.13 MB)\n","
\n"]},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m23,588,480\u001b[0m (89.98 MB)\n"],"text/html":["
 Non-trainable params: 23,588,480 (89.98 MB)\n","
\n"]},"metadata":{}}]},{"cell_type":"code","source":["# RESNET50 - STAGE 1 TRAINING\n","\n","batch_size = 20\n","epochs = 20\n","\n","steps = len(train_paths) // batch_size\n","\n","\n","# Early stopping\n","early_stopping = tf.keras.callbacks.EarlyStopping(\n"," monitor='val_loss',\n"," patience=5,\n"," restore_best_weights=True,\n"," verbose=1\n",")\n","\n","\n","# Reduce learning rate when validation loss stops improving\n","reduce_lr = tf.keras.callbacks.ReduceLROnPlateau(\n"," monitor='val_loss',\n"," factor=0.2,\n"," patience=2,\n"," min_lr=1e-7,\n"," verbose=1\n",")\n","\n","\n","# Save best ResNet50 model\n","checkpoint = tf.keras.callbacks.ModelCheckpoint(\n"," 'resnet50_stage1_best.keras',\n"," monitor='val_loss',\n"," save_best_only=True,\n"," verbose=1\n",")\n","\n","\n","# Train\n","history = model.fit(\n"," datagen(\n"," train_paths,\n"," train_labels,\n"," batch_size=batch_size,\n"," epochs=epochs\n"," ),\n","\n"," epochs=epochs,\n","\n"," steps_per_epoch=steps,\n","\n"," validation_data=datagen(\n"," val_paths,\n"," val_labels,\n"," batch_size=batch_size,\n"," epochs=1\n"," ),\n","\n"," validation_steps=len(val_paths) // batch_size,\n","\n"," callbacks=[\n"," early_stopping,\n"," reduce_lr,\n"," checkpoint\n"," ]\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Z4-uiyDlKnUm","executionInfo":{"status":"ok","timestamp":1787460044820,"user_tz":-330,"elapsed":2588959,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"5d8f9589-8ce5-4093-b5ea-71bc23549278"},"execution_count":20,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8s/step - loss: 1.5783 - sparse_categorical_accuracy: 0.3982\n","Epoch 1: val_loss improved from None to 1.32247, saving model to resnet50_stage1_best.keras\n","\n","Epoch 1: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2230s\u001b[0m 9s/step - loss: 1.3952 - sparse_categorical_accuracy: 0.4595 - val_loss: 1.3225 - val_sparse_categorical_accuracy: 0.3798 - learning_rate: 1.0000e-04\n","Epoch 2/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 84ms/step - loss: 1.1118 - sparse_categorical_accuracy: 0.5565"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.13/dist-packages/keras/src/trainers/epoch_iterator.py:164: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n"," self._interrupted_warning()\n"]},{"output_type":"stream","name":"stdout","text":["\n","Epoch 2: val_loss improved from 1.32247 to 0.94244, saving model to resnet50_stage1_best.keras\n","\n","Epoch 2: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 91ms/step - loss: 1.0841 - sparse_categorical_accuracy: 0.5685 - val_loss: 0.9424 - val_sparse_categorical_accuracy: 0.6000 - learning_rate: 1.0000e-04\n","Epoch 3/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 79ms/step - loss: 1.0201 - sparse_categorical_accuracy: 0.6104\n","Epoch 3: val_loss improved from 0.94244 to 0.69368, saving model to resnet50_stage1_best.keras\n","\n","Epoch 3: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 85ms/step - loss: 0.9941 - sparse_categorical_accuracy: 0.6120 - val_loss: 0.6937 - val_sparse_categorical_accuracy: 0.7250 - learning_rate: 1.0000e-04\n","Epoch 4/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 83ms/step - loss: 0.9714 - sparse_categorical_accuracy: 0.6142\n","Epoch 4: val_loss improved from 0.69368 to 0.61242, saving model to resnet50_stage1_best.keras\n","\n","Epoch 4: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 90ms/step - loss: 0.9267 - sparse_categorical_accuracy: 0.6335 - val_loss: 0.6124 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 1.0000e-04\n","Epoch 5/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 77ms/step - loss: 0.8975 - sparse_categorical_accuracy: 0.6402\n","Epoch 5: val_loss did not improve from 0.61242\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 78ms/step - loss: 0.8833 - sparse_categorical_accuracy: 0.6470 - val_loss: 0.7027 - val_sparse_categorical_accuracy: 0.7750 - learning_rate: 1.0000e-04\n","Epoch 6/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 83ms/step - loss: 0.8420 - sparse_categorical_accuracy: 0.6664\n","Epoch 6: val_loss improved from 0.61242 to 0.57251, saving model to resnet50_stage1_best.keras\n","\n","Epoch 6: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 93ms/step - loss: 0.8334 - sparse_categorical_accuracy: 0.6669 - val_loss: 0.5725 - val_sparse_categorical_accuracy: 0.7500 - learning_rate: 1.0000e-04\n","Epoch 7/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 77ms/step - loss: 0.8218 - sparse_categorical_accuracy: 0.6708\n","Epoch 7: val_loss did not improve from 0.57251\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 78ms/step - loss: 0.8097 - sparse_categorical_accuracy: 0.6758 - val_loss: 0.7744 - val_sparse_categorical_accuracy: 0.7250 - learning_rate: 1.0000e-04\n","Epoch 8/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 81ms/step - loss: 0.8063 - sparse_categorical_accuracy: 0.6723\n","Epoch 8: ReduceLROnPlateau reducing learning rate to 1.9999999494757503e-05.\n","\n","Epoch 8: val_loss did not improve from 0.57251\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 81ms/step - loss: 0.7736 - sparse_categorical_accuracy: 0.6850 - val_loss: 0.7017 - val_sparse_categorical_accuracy: 0.7250 - learning_rate: 1.0000e-04\n","Epoch 9/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 85ms/step - loss: 0.7827 - sparse_categorical_accuracy: 0.6861\n","Epoch 9: val_loss improved from 0.57251 to 0.52324, saving model to resnet50_stage1_best.keras\n","\n","Epoch 9: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 91ms/step - loss: 0.7642 - sparse_categorical_accuracy: 0.6972 - val_loss: 0.5232 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 2.0000e-05\n","Epoch 10/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 77ms/step - loss: 0.7537 - sparse_categorical_accuracy: 0.6982\n","Epoch 10: val_loss did not improve from 0.52324\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 77ms/step - loss: 0.7439 - sparse_categorical_accuracy: 0.7010 - val_loss: 0.5309 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 2.0000e-05\n","Epoch 11/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 82ms/step - loss: 0.7821 - sparse_categorical_accuracy: 0.6975\n","Epoch 11: ReduceLROnPlateau reducing learning rate to 3.999999898951501e-06.\n","\n","Epoch 11: val_loss did not improve from 0.52324\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 83ms/step - loss: 0.7618 - sparse_categorical_accuracy: 0.7036 - val_loss: 0.5370 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 2.0000e-05\n","Epoch 12/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 81ms/step - loss: 0.7687 - sparse_categorical_accuracy: 0.7034\n","Epoch 12: val_loss improved from 0.52324 to 0.52060, saving model to resnet50_stage1_best.keras\n","\n","Epoch 12: finished saving model to resnet50_stage1_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 88ms/step - loss: 0.7490 - sparse_categorical_accuracy: 0.7023 - val_loss: 0.5206 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 4.0000e-06\n","Epoch 13/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 76ms/step - loss: 0.7517 - sparse_categorical_accuracy: 0.7113\n","Epoch 13: val_loss did not improve from 0.52060\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 77ms/step - loss: 0.7370 - sparse_categorical_accuracy: 0.7132 - val_loss: 0.5241 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 4.0000e-06\n","Epoch 14/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 85ms/step - loss: 0.7725 - sparse_categorical_accuracy: 0.6947\n","Epoch 14: ReduceLROnPlateau reducing learning rate to 7.999999979801942e-07.\n","\n","Epoch 14: val_loss did not improve from 0.52060\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 86ms/step - loss: 0.7581 - sparse_categorical_accuracy: 0.6979 - val_loss: 0.5271 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 4.0000e-06\n","Epoch 15/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 82ms/step - loss: 0.7677 - sparse_categorical_accuracy: 0.6996\n","Epoch 15: val_loss did not improve from 0.52060\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 83ms/step - loss: 0.7462 - sparse_categorical_accuracy: 0.7004 - val_loss: 0.5238 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 8.0000e-07\n","Epoch 16/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 83ms/step - loss: 0.7623 - sparse_categorical_accuracy: 0.6969\n","Epoch 16: ReduceLROnPlateau reducing learning rate to 1.600000018697756e-07.\n","\n","Epoch 16: val_loss did not improve from 0.52060\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 84ms/step - loss: 0.7447 - sparse_categorical_accuracy: 0.7059 - val_loss: 0.5222 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 8.0000e-07\n","Epoch 17/20\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 82ms/step - loss: 0.7519 - sparse_categorical_accuracy: 0.7026\n","Epoch 17: val_loss did not improve from 0.52060\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 83ms/step - loss: 0.7430 - sparse_categorical_accuracy: 0.7063 - val_loss: 0.5220 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 1.6000e-07\n","Epoch 17: early stopping\n","Restoring model weights from the end of the best epoch: 12.\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"CPof__kiNy-j"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["print(\"Best validation accuracy:\")\n","print(max(history.history['val_sparse_categorical_accuracy']))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Ig4JaLoGXLxw","executionInfo":{"status":"ok","timestamp":1787461842115,"user_tz":-330,"elapsed":34,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"61a44ab6-6a5a-4b37-bea3-bcec1d30859d"},"execution_count":21,"outputs":[{"output_type":"stream","name":"stdout","text":["Best validation accuracy:\n","0.8500000238418579\n"]}]},{"cell_type":"code","source":["print(\"Best validation loss:\")\n","print(min(history.history['val_loss']))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TXBzbim0XPyn","executionInfo":{"status":"ok","timestamp":1787461843973,"user_tz":-330,"elapsed":46,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"15123d1b-0fc3-4e96-835f-3dbd83e6d5ef"},"execution_count":22,"outputs":[{"output_type":"stream","name":"stdout","text":["Best validation loss:\n","0.5206043720245361\n"]}]},{"cell_type":"code","source":["plt.figure(figsize=(10,5))\n","plt.grid(True)\n","plt.plot(history.history['sparse_categorical_accuracy'], '.g-', linewidth=2)\n","plt.plot(history.history['loss'], '.r-', linewidth=2.5)\n","plt.title('Model Training History')\n","plt.xlabel('Epoch')\n","plt.ylabel('Value')\n","plt.xticks([x for x in range(epochs)])\n","plt.legend(['accuracy', 'loss'], loc='upper left', bbox_to_anchor=(1, 1))\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"NoGrm8BxWI_X","executionInfo":{"status":"ok","timestamp":1787461845935,"user_tz":-330,"elapsed":319,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"e0aa2816-5305-4d7d-e102-aaee5015c911"},"execution_count":23,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["import matplotlib.pyplot as plt\n","\n","plt.plot(history.history['sparse_categorical_accuracy'], label='Training')\n","plt.plot(history.history['val_sparse_categorical_accuracy'], label='Validation')\n","\n","plt.xlabel('Epoch')\n","plt.ylabel('Accuracy')\n","plt.legend()\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":449},"id":"H5uHADMeXVUh","executionInfo":{"status":"ok","timestamp":1787461846853,"user_tz":-330,"elapsed":133,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"ac040a2d-9a80-4224-953d-babcc5905c54"},"execution_count":24,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":[],"metadata":{"id":"skRQtWRQVy2d","executionInfo":{"status":"ok","timestamp":1787461847890,"user_tz":-330,"elapsed":3,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":24,"outputs":[]},{"cell_type":"code","source":["test_images = open_images(test_paths) #load and augment test images\n","test_labels_encoded = encode_label(test_labels) # encode the test labels\n","#Predict on test data"],"metadata":{"id":"C9Al8JIrXxsb","executionInfo":{"status":"ok","timestamp":1787462425883,"user_tz":-330,"elapsed":577345,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":25,"outputs":[]},{"cell_type":"code","source":["test_predictions = model.predict(test_images)\n","#predict using the trained model"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"PQ1XNQt1XyZL","executionInfo":{"status":"ok","timestamp":1787462433469,"user_tz":-330,"elapsed":7582,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"c06b5ea5-d8c8-4baf-ad41-3de9809b5ad3"},"execution_count":26,"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 31ms/step\n"]}]},{"cell_type":"code","source":["print(classification_report(test_labels_encoded, np.argmax(test_predictions, axis=1)))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Ksy834n4YDX-","executionInfo":{"status":"ok","timestamp":1787462577391,"user_tz":-330,"elapsed":52,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"47fe0ad1-c335-458f-f772-8de219c7a788"},"execution_count":27,"outputs":[{"output_type":"stream","name":"stdout","text":[" precision recall f1-score support\n","\n"," 0 0.71 0.86 0.78 400\n"," 1 0.73 0.86 0.79 400\n"," 2 0.61 0.35 0.45 400\n"," 3 0.62 0.62 0.62 400\n","\n"," accuracy 0.68 1600\n"," macro avg 0.66 0.68 0.66 1600\n","weighted avg 0.66 0.68 0.66 1600\n","\n"]}]},{"cell_type":"code","source":["conf_matrix = confusion_matrix(test_labels_encoded, np.argmax(test_predictions, axis=1))\n","print(conf_matrix)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GB4e4zE0YDNl","executionInfo":{"status":"ok","timestamp":1787462577796,"user_tz":-330,"elapsed":18,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"58ba3761-822b-4592-88e6-fb9918fdc377"},"execution_count":28,"outputs":[{"output_type":"stream","name":"stdout","text":["[[345 7 19 29]\n"," [ 22 345 12 21]\n"," [ 85 69 141 105]\n"," [ 37 54 60 249]]\n"]}]},{"cell_type":"markdown","source":["# **Stage 2 fine-tuning**"],"metadata":{"id":"bRRjOC6rOKP4"}},{"cell_type":"code","source":["# STAGE 2 - FINE TUNING RESNET50\n","\n","# Freeze everything first\n","for layer in base_model.layers:\n"," layer.trainable = False\n","\n","\n","# Unfreeze the final convolutional layers\n","# ResNet50 uses names such as conv5_block1, conv5_block2, etc.\n","\n","set_trainable = False\n","\n","for layer in base_model.layers:\n","\n"," if layer.name == 'conv5_block1_1_conv':\n"," set_trainable = True\n","\n"," if set_trainable:\n"," layer.trainable = True\n","\n","\n","# Check which layers are trainable\n","print(\"\\nTrainable ResNet50 layers:\")\n","\n","for layer in base_model.layers:\n"," print(layer.name, layer.trainable)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yDDIItDxOJR_","executionInfo":{"status":"ok","timestamp":1787462649319,"user_tz":-330,"elapsed":52,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"14eb5307-031a-4370-831c-f63a5531dc83"},"execution_count":29,"outputs":[{"output_type":"stream","name":"stdout","text":["\n","Trainable ResNet50 layers:\n","input_layer False\n","conv1_pad False\n","conv1_conv False\n","conv1_bn False\n","conv1_relu False\n","pool1_pad False\n","pool1_pool False\n","conv2_block1_1_conv False\n","conv2_block1_1_bn False\n","conv2_block1_1_relu False\n","conv2_block1_2_conv False\n","conv2_block1_2_bn False\n","conv2_block1_2_relu False\n","conv2_block1_0_conv False\n","conv2_block1_3_conv False\n","conv2_block1_0_bn False\n","conv2_block1_3_bn False\n","conv2_block1_add False\n","conv2_block1_out False\n","conv2_block2_1_conv False\n","conv2_block2_1_bn False\n","conv2_block2_1_relu False\n","conv2_block2_2_conv False\n","conv2_block2_2_bn False\n","conv2_block2_2_relu False\n","conv2_block2_3_conv False\n","conv2_block2_3_bn False\n","conv2_block2_add False\n","conv2_block2_out False\n","conv2_block3_1_conv False\n","conv2_block3_1_bn False\n","conv2_block3_1_relu False\n","conv2_block3_2_conv False\n","conv2_block3_2_bn False\n","conv2_block3_2_relu False\n","conv2_block3_3_conv False\n","conv2_block3_3_bn False\n","conv2_block3_add False\n","conv2_block3_out False\n","conv3_block1_1_conv False\n","conv3_block1_1_bn False\n","conv3_block1_1_relu False\n","conv3_block1_2_conv False\n","conv3_block1_2_bn False\n","conv3_block1_2_relu False\n","conv3_block1_0_conv False\n","conv3_block1_3_conv False\n","conv3_block1_0_bn False\n","conv3_block1_3_bn False\n","conv3_block1_add False\n","conv3_block1_out False\n","conv3_block2_1_conv False\n","conv3_block2_1_bn False\n","conv3_block2_1_relu False\n","conv3_block2_2_conv False\n","conv3_block2_2_bn False\n","conv3_block2_2_relu False\n","conv3_block2_3_conv False\n","conv3_block2_3_bn False\n","conv3_block2_add False\n","conv3_block2_out False\n","conv3_block3_1_conv False\n","conv3_block3_1_bn False\n","conv3_block3_1_relu False\n","conv3_block3_2_conv False\n","conv3_block3_2_bn False\n","conv3_block3_2_relu False\n","conv3_block3_3_conv False\n","conv3_block3_3_bn False\n","conv3_block3_add False\n","conv3_block3_out False\n","conv3_block4_1_conv False\n","conv3_block4_1_bn False\n","conv3_block4_1_relu False\n","conv3_block4_2_conv False\n","conv3_block4_2_bn False\n","conv3_block4_2_relu False\n","conv3_block4_3_conv False\n","conv3_block4_3_bn False\n","conv3_block4_add False\n","conv3_block4_out False\n","conv4_block1_1_conv False\n","conv4_block1_1_bn False\n","conv4_block1_1_relu False\n","conv4_block1_2_conv False\n","conv4_block1_2_bn False\n","conv4_block1_2_relu False\n","conv4_block1_0_conv False\n","conv4_block1_3_conv False\n","conv4_block1_0_bn False\n","conv4_block1_3_bn False\n","conv4_block1_add False\n","conv4_block1_out False\n","conv4_block2_1_conv False\n","conv4_block2_1_bn False\n","conv4_block2_1_relu False\n","conv4_block2_2_conv False\n","conv4_block2_2_bn False\n","conv4_block2_2_relu False\n","conv4_block2_3_conv False\n","conv4_block2_3_bn False\n","conv4_block2_add False\n","conv4_block2_out False\n","conv4_block3_1_conv False\n","conv4_block3_1_bn False\n","conv4_block3_1_relu False\n","conv4_block3_2_conv False\n","conv4_block3_2_bn False\n","conv4_block3_2_relu False\n","conv4_block3_3_conv False\n","conv4_block3_3_bn False\n","conv4_block3_add False\n","conv4_block3_out False\n","conv4_block4_1_conv False\n","conv4_block4_1_bn False\n","conv4_block4_1_relu False\n","conv4_block4_2_conv False\n","conv4_block4_2_bn False\n","conv4_block4_2_relu False\n","conv4_block4_3_conv False\n","conv4_block4_3_bn False\n","conv4_block4_add False\n","conv4_block4_out False\n","conv4_block5_1_conv False\n","conv4_block5_1_bn False\n","conv4_block5_1_relu False\n","conv4_block5_2_conv False\n","conv4_block5_2_bn False\n","conv4_block5_2_relu False\n","conv4_block5_3_conv False\n","conv4_block5_3_bn False\n","conv4_block5_add False\n","conv4_block5_out False\n","conv4_block6_1_conv False\n","conv4_block6_1_bn False\n","conv4_block6_1_relu False\n","conv4_block6_2_conv False\n","conv4_block6_2_bn False\n","conv4_block6_2_relu False\n","conv4_block6_3_conv False\n","conv4_block6_3_bn False\n","conv4_block6_add False\n","conv4_block6_out False\n","conv5_block1_1_conv True\n","conv5_block1_1_bn True\n","conv5_block1_1_relu True\n","conv5_block1_2_conv True\n","conv5_block1_2_bn True\n","conv5_block1_2_relu True\n","conv5_block1_0_conv True\n","conv5_block1_3_conv True\n","conv5_block1_0_bn True\n","conv5_block1_3_bn True\n","conv5_block1_add True\n","conv5_block1_out True\n","conv5_block2_1_conv True\n","conv5_block2_1_bn True\n","conv5_block2_1_relu True\n","conv5_block2_2_conv True\n","conv5_block2_2_bn True\n","conv5_block2_2_relu True\n","conv5_block2_3_conv True\n","conv5_block2_3_bn True\n","conv5_block2_add True\n","conv5_block2_out True\n","conv5_block3_1_conv True\n","conv5_block3_1_bn True\n","conv5_block3_1_relu True\n","conv5_block3_2_conv True\n","conv5_block3_2_bn True\n","conv5_block3_2_relu True\n","conv5_block3_3_conv True\n","conv5_block3_3_bn True\n","conv5_block3_add True\n","conv5_block3_out True\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"CIIbG4Y-Oh0R"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# COMPILE FOR RESNET50 FINE-TUNING\n","\n","model.compile(\n"," optimizer=Adam(learning_rate=1e-5),\n"," loss='sparse_categorical_crossentropy',\n"," metrics=['sparse_categorical_accuracy']\n",")"],"metadata":{"id":"rxJy5ulQOmDi","executionInfo":{"status":"ok","timestamp":1787462681027,"user_tz":-330,"elapsed":4,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":30,"outputs":[]},{"cell_type":"code","source":["# FINE-TUNE RESNET50\n","\n","fine_tune_epochs = 15\n","\n","fine_tune_callbacks = [\n","\n"," tf.keras.callbacks.EarlyStopping(\n"," monitor='val_loss',\n"," patience=5,\n"," restore_best_weights=True,\n"," verbose=1\n"," ),\n","\n"," tf.keras.callbacks.ReduceLROnPlateau(\n"," monitor='val_loss',\n"," factor=0.2,\n"," patience=2,\n"," min_lr=1e-8,\n"," verbose=1\n"," ),\n","\n"," tf.keras.callbacks.ModelCheckpoint(\n"," 'resnet50_finetuned_best.keras',\n"," monitor='val_loss',\n"," save_best_only=True,\n"," verbose=1\n"," )\n","]\n","\n","\n","history_finetune = model.fit(\n","\n"," datagen(\n"," train_paths,\n"," train_labels,\n"," batch_size=batch_size,\n"," epochs=fine_tune_epochs\n"," ),\n","\n"," epochs=fine_tune_epochs,\n","\n"," steps_per_epoch=len(train_paths) // batch_size,\n","\n"," validation_data=datagen(\n"," val_paths,\n"," val_labels,\n"," batch_size=batch_size,\n"," epochs=1\n"," ),\n","\n"," validation_steps=len(val_paths) // batch_size,\n","\n"," callbacks=fine_tune_callbacks\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GIajgbqcOpLC","executionInfo":{"status":"ok","timestamp":1787463074617,"user_tz":-330,"elapsed":352882,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"2d14ddbe-3c30-475f-8c73-64272a739cf4"},"execution_count":32,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 87ms/step - loss: 1.2413 - sparse_categorical_accuracy: 0.5144\n","Epoch 1: val_loss improved from None to 1.20781, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 1: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━��━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 130ms/step - loss: 1.0849 - sparse_categorical_accuracy: 0.5779 - val_loss: 1.2078 - val_sparse_categorical_accuracy: 0.4905 - learning_rate: 1.0000e-05\n","Epoch 2/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 86ms/step - loss: 0.8428 - sparse_categorical_accuracy: 0.6734"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.13/dist-packages/keras/src/trainers/epoch_iterator.py:164: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n"," self._interrupted_warning()\n"]},{"output_type":"stream","name":"stdout","text":["\n","Epoch 2: val_loss improved from 1.20781 to 0.69639, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 2: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 94ms/step - loss: 0.8103 - sparse_categorical_accuracy: 0.6924 - val_loss: 0.6964 - val_sparse_categorical_accuracy: 0.6500 - learning_rate: 1.0000e-05\n","Epoch 3/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 93ms/step - loss: 0.7784 - sparse_categorical_accuracy: 0.7057\n","Epoch 3: val_loss improved from 0.69639 to 0.60638, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 3: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m24s\u001b[0m 102ms/step - loss: 0.7481 - sparse_categorical_accuracy: 0.7164 - val_loss: 0.6064 - val_sparse_categorical_accuracy: 0.7500 - learning_rate: 1.0000e-05\n","Epoch 4/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 88ms/step - loss: 0.6946 - sparse_categorical_accuracy: 0.7365\n","Epoch 4: val_loss improved from 0.60638 to 0.59147, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 4: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m24s\u001b[0m 101ms/step - loss: 0.6825 - sparse_categorical_accuracy: 0.7412 - val_loss: 0.5915 - val_sparse_categorical_accuracy: 0.7250 - learning_rate: 1.0000e-05\n","Epoch 5/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 82ms/step - loss: 0.6561 - sparse_categorical_accuracy: 0.7473\n","Epoch 5: val_loss improved from 0.59147 to 0.47099, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 5: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 93ms/step - loss: 0.6442 - sparse_categorical_accuracy: 0.7522 - val_loss: 0.4710 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 1.0000e-05\n","Epoch 6/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 88ms/step - loss: 0.6440 - sparse_categorical_accuracy: 0.7608\n","Epoch 6: val_loss did not improve from 0.47099\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 89ms/step - loss: 0.6107 - sparse_categorical_accuracy: 0.7734 - val_loss: 0.4753 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 1.0000e-05\n","Epoch 7/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 87ms/step - loss: 0.5616 - sparse_categorical_accuracy: 0.7938\n","Epoch 7: ReduceLROnPlateau reducing learning rate to 1.9999999494757505e-06.\n","\n","Epoch 7: val_loss did not improve from 0.47099\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 87ms/step - loss: 0.5572 - sparse_categorical_accuracy: 0.7890 - val_loss: 0.5529 - val_sparse_categorical_accuracy: 0.8000 - learning_rate: 1.0000e-05\n","Epoch 8/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 92ms/step - loss: 0.5141 - sparse_categorical_accuracy: 0.8031\n","Epoch 8: val_loss did not improve from 0.47099\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━���━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 92ms/step - loss: 0.5005 - sparse_categorical_accuracy: 0.8102 - val_loss: 0.4728 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 2.0000e-06\n","Epoch 9/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 89ms/step - loss: 0.4862 - sparse_categorical_accuracy: 0.8134\n","Epoch 9: val_loss improved from 0.47099 to 0.43664, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 9: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 99ms/step - loss: 0.4602 - sparse_categorical_accuracy: 0.8228 - val_loss: 0.4366 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 2.0000e-06\n","Epoch 10/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 86ms/step - loss: 0.4503 - sparse_categorical_accuracy: 0.8297\n","Epoch 10: val_loss improved from 0.43664 to 0.42154, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 10: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 96ms/step - loss: 0.4391 - sparse_categorical_accuracy: 0.8374 - val_loss: 0.4215 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 2.0000e-06\n","Epoch 11/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 81ms/step - loss: 0.4263 - sparse_categorical_accuracy: 0.8417\n","Epoch 11: val_loss improved from 0.42154 to 0.39053, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 11: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 92ms/step - loss: 0.4157 - sparse_categorical_accuracy: 0.8399 - val_loss: 0.3905 - val_sparse_categorical_accuracy: 0.8500 - learning_rate: 2.0000e-06\n","Epoch 12/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 86ms/step - loss: 0.4476 - sparse_categorical_accuracy: 0.8273\n","Epoch 12: val_loss improved from 0.39053 to 0.37626, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 12: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 95ms/step - loss: 0.4179 - sparse_categorical_accuracy: 0.8380 - val_loss: 0.3763 - val_sparse_categorical_accuracy: 0.8250 - learning_rate: 2.0000e-06\n","Epoch 13/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 82ms/step - loss: 0.4187 - sparse_categorical_accuracy: 0.8356\n","Epoch 13: val_loss improved from 0.37626 to 0.36290, saving model to resnet50_finetuned_best.keras\n","\n","Epoch 13: finished saving model to resnet50_finetuned_best.keras\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 97ms/step - loss: 0.3933 - sparse_categorical_accuracy: 0.8477 - val_loss: 0.3629 - val_sparse_categorical_accuracy: 0.8750 - learning_rate: 2.0000e-06\n","Epoch 14/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 86ms/step - loss: 0.3983 - sparse_categorical_accuracy: 0.8499\n","Epoch 14: val_loss did not improve from 0.36290\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 87ms/step - loss: 0.3717 - sparse_categorical_accuracy: 0.8597 - val_loss: 0.3786 - val_sparse_categorical_accuracy: 0.9250 - learning_rate: 2.0000e-06\n","Epoch 15/15\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 87ms/step - loss: 0.3806 - sparse_categorical_accuracy: 0.8578\n","Epoch 15: ReduceLROnPlateau reducing learning rate to 3.999999989900971e-07.\n","\n","Epoch 15: val_loss did not improve from 0.36290\n","\u001b[1m238/238\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 88ms/step - loss: 0.3582 - sparse_categorical_accuracy: 0.8683 - val_loss: 0.3806 - val_sparse_categorical_accuracy: 0.9000 - learning_rate: 2.0000e-06\n","Restoring model weights from the end of the best epoch: 13.\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"NmIPmXqjaXhu"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["**Test**"],"metadata":{"id":"u1Yxil_2cHKC"}},{"cell_type":"code","source":["def load_test_images(paths):\n"," images = []\n","\n"," for path in paths:\n"," image = load_img(\n"," path,\n"," target_size=(IMAGE_SIZE, IMAGE_SIZE)\n"," )\n","\n"," image = np.array(image) / 255.0\n"," images.append(image)\n","\n"," return np.array(images)"],"metadata":{"id":"cJTtX_o4cGzK","executionInfo":{"status":"ok","timestamp":1787463074665,"user_tz":-330,"elapsed":2,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":33,"outputs":[]},{"cell_type":"code","source":["# Load test images WITHOUT augmentation\n","X_test = load_test_images(test_paths)\n","\n","# Encode test labels\n","y_test = encode_label(test_labels)\n","\n","print(\"Test images:\", X_test.shape)\n","print(\"Test labels:\", y_test.shape)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"uWpwj6hHcJq6","executionInfo":{"status":"ok","timestamp":1787463079872,"user_tz":-330,"elapsed":5208,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"612cb727-1d13-4ad5-ad6a-59ed0ff74fa3"},"execution_count":34,"outputs":[{"output_type":"stream","name":"stdout","text":["Test images: (1600, 128, 128, 3)\n","Test labels: (1600,)\n"]}]},{"cell_type":"code","source":["y_pred_prob = model.predict(X_test)\n","\n","y_pred = np.argmax(y_pred_prob, axis=1)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"QQHcTbH0cL0b","executionInfo":{"status":"ok","timestamp":1787463086281,"user_tz":-330,"elapsed":6408,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"aec7031a-2502-4202-df30-7671b439c8ae"},"execution_count":35,"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 33ms/step\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"gXEt14IvcQ-7","executionInfo":{"status":"ok","timestamp":1787463086308,"user_tz":-330,"elapsed":26,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":35,"outputs":[]},{"cell_type":"code","source":["from sklearn.metrics import confusion_matrix\n","import seaborn as sns\n","import matplotlib.pyplot as plt\n","\n","cm = confusion_matrix(y_test, y_pred)\n","\n","print(cm)\n","\n","plt.figure(figsize=(7, 6))\n","\n","sns.heatmap(\n"," cm,\n"," annot=True,\n"," fmt='d',\n"," cmap='Blues'\n",")\n","\n","plt.xlabel(\"Predicted\")\n","plt.ylabel(\"Actual\")\n","plt.title(\"Confusion Matrix - Fine-Tuned VGG16\")\n","\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":633},"id":"XjoV2kUqcTJr","executionInfo":{"status":"ok","timestamp":1787463086489,"user_tz":-330,"elapsed":180,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"dc213a7d-a651-46aa-a90b-8edaa246e4c1"},"execution_count":36,"outputs":[{"output_type":"stream","name":"stdout","text":["[[362 3 18 17]\n"," [ 2 395 2 1]\n"," [ 31 41 264 64]\n"," [ 10 65 53 272]]\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["from sklearn.metrics import classification_report\n","\n","class_names = sorted(os.listdir(train_dir))\n","\n","print(\n"," classification_report(\n"," y_test,\n"," y_pred,\n"," target_names=class_names\n"," )\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"101ZR_CPcZeM","executionInfo":{"status":"ok","timestamp":1787463086498,"user_tz":-330,"elapsed":8,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"19bf5c75-d792-4eed-a207-558d31d85fa7"},"execution_count":37,"outputs":[{"output_type":"stream","name":"stdout","text":[" precision recall f1-score support\n","\n"," glioma 0.89 0.91 0.90 400\n"," meningioma 0.78 0.99 0.87 400\n"," notumor 0.78 0.66 0.72 400\n"," pituitary 0.77 0.68 0.72 400\n","\n"," accuracy 0.81 1600\n"," macro avg 0.81 0.81 0.80 1600\n","weighted avg 0.81 0.81 0.80 1600\n","\n"]}]},{"cell_type":"code","source":["from sklearn.metrics import accuracy_score\n","\n","accuracy = accuracy_score(y_test, y_pred)\n","\n","print(f\"Test Accuracy: {accuracy:.4f}\")\n","print(f\"Test Accuracy: {accuracy * 100:.2f}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"T6GbXqQMccuE","executionInfo":{"status":"ok","timestamp":1787463086506,"user_tz":-330,"elapsed":7,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"4bd15108-1db5-4b2d-bd15-14bc33ca6771"},"execution_count":38,"outputs":[{"output_type":"stream","name":"stdout","text":["Test Accuracy: 0.8081\n","Test Accuracy: 80.81%\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"WwjdM_9Bcful","executionInfo":{"status":"ok","timestamp":1787463086507,"user_tz":-330,"elapsed":1,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":38,"outputs":[]},{"cell_type":"code","source":["import numpy as np\n","import tensorflow as tf\n","from tensorflow.keras.models import load_model\n","from sklearn.metrics import accuracy_score\n","\n","# LOAD BOTH MODELS\n","\n","# Original VGG16 model\n","original_model = load_model(\"/content/drive/MyDrive/cancer-cnn/resnet50_stage1_best.keras\")\n","\n","# Fine-tuned VGG16 model\n","finetuned_model = load_model(\"/content/drive/MyDrive/cancer-cnn/resnet50_finetuned_best.keras\")\n","\n","# LOAD TEST DATA\n","X_test = load_test_images(test_paths)\n","y_test = encode_label(test_labels)\n","\n","# PREDICTIONS - ORIGINAL MODEL\n","original_prob = original_model.predict(\n"," X_test,\n"," verbose=1\n",")\n","\n","original_pred = np.argmax(\n"," original_prob,\n"," axis=1\n",")\n","\n","original_accuracy = accuracy_score(\n"," y_test,\n"," original_pred\n",")\n","\n","\n","# PREDICTIONS - FINE-TUNED MODEL\n","finetuned_prob = finetuned_model.predict(\n"," X_test,\n"," verbose=1\n",")\n","finetuned_pred = np.argmax(\n"," finetuned_prob,\n"," axis=1\n",")\n","finetuned_accuracy = accuracy_score(\n"," y_test,\n"," finetuned_pred\n",")\n","\n","\n","# PRINT RESULTS\n","print(\" MODEL ACCURACY COMPARISON\")\n","print(\" \")\n","\n","print(\n"," f\"Original VGG16 : \"\n"," f\"{original_accuracy * 100:.2f}%\"\n",")\n","\n","print(\n"," f\"Fine-tuned VGG16 : \"\n"," f\"{finetuned_accuracy * 100:.2f}%\"\n",")\n","\n","print(\" \")\n","\n","improvement = (\n"," finetuned_accuracy - original_accuracy\n",") * 100\n","\n","print(\n"," f\"Improvement : \"\n"," f\"{improvement:+.2f} percentage points\"\n",")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FUzxK7bvdLKC","executionInfo":{"status":"ok","timestamp":1787463522498,"user_tz":-330,"elapsed":55716,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"55e6327f-f2dd-45fa-eac6-3fb333171fe0"},"execution_count":40,"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 34ms/step\n","\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 38ms/step\n"," MODEL ACCURACY COMPARISON\n"," \n","Original VGG16 : 8.12%\n","Fine-tuned VGG16 : 4.38%\n"," \n","Improvement : -3.75 percentage points\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"XOW2SUZ4fMqH","executionInfo":{"status":"aborted","timestamp":1787463086553,"user_tz":-330,"elapsed":3,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["**prediction**"],"metadata":{"id":"V1CJw3VpneEK"}},{"cell_type":"code","source":["from tensorflow.keras.models import load_model\n","model = load_model('/content/drive/MyDrive/cancer-cnn/resnet50_finetuned_best.keras')"],"metadata":{"id":"QOGO9Cg4fYaR","executionInfo":{"status":"ok","timestamp":1787463524688,"user_tz":-330,"elapsed":2186,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":41,"outputs":[]},{"cell_type":"code","source":["# IMPORTANT:\n","# This mapping matches the output you observed:\n","# model output 0 -> pituitary\n","# model output 1 -> notumor\n","# model output -> meningioma\n","# model output 3 -> glioma\n","\n","\n","class_names = [\n"," \"pituitary\",\n"," \"notumor\",\n"," \"meningioma\",\n"," \"glioma\"\n","]"],"metadata":{"id":"wo8iwPFOfYXu","executionInfo":{"status":"ok","timestamp":1787463524695,"user_tz":-330,"elapsed":5,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":42,"outputs":[]},{"cell_type":"code","source":["import matplotlib.pyplot as plt\n","from tensorflow.keras.utils import load_img, img_to_array\n","import numpy as np\n","\n","def predict_image(image_path, model):\n","\n"," # Load image\n"," image = load_img(\n"," image_path,\n"," target_size=(IMAGE_SIZE, IMAGE_SIZE)\n"," )\n","\n"," # Convert image to array\n"," image_array = img_to_array(image)\n","\n"," # Normalize\n"," image_array = image_array / 255.0\n","\n"," # Add batch dimension\n"," input_image = np.expand_dims(image_array, axis=0)\n","\n"," # Predict\n"," prediction = model.predict(input_image, verbose=0)[0]\n","\n"," predicted_index = np.argmax(prediction)\n"," predicted_class = class_names[predicted_index]\n"," confidence = prediction[predicted_index] * 100\n","\n"," # Display image\n"," plt.figure(figsize=(6, 6))\n"," plt.imshow(image)\n"," plt.axis(\"off\")\n","\n"," plt.title(\n"," f\"Prediction: {predicted_class}\\n\"\n"," f\"Confidence: {confidence:.2f}%\"\n"," )\n","\n"," plt.show()\n","\n"," # Print probabilities\n"," print(\"MRI PREDICTION\")\n"," print(\" \")\n","\n"," for i, class_name in enumerate(class_names):\n"," print(\n"," f\"{class_name:12s}: \"\n"," f\"{prediction[i] * 100:.2f}%\"\n"," )\n","\n"," print(\" \")\n"," print(f\"Prediction : {predicted_class}\")\n"," print(f\"Confidence : {confidence:.2f}%\")\n","\n"," return predicted_class, confidence"],"metadata":{"id":"rOBg6snommAE","executionInfo":{"status":"ok","timestamp":1787463524695,"user_tz":-330,"elapsed":4,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":43,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"qnPgpAXKmlzu","executionInfo":{"status":"ok","timestamp":1787463524696,"user_tz":-330,"elapsed":4,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":43,"outputs":[]},{"cell_type":"code","source":["predict_image(\n"," \"/content/drive/MyDrive/cancer-cnn/extracted data/Testing/glioma/Te-gl_131.jpg\",\n"," model\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":716},"id":"Jd3h7DCZfMgx","executionInfo":{"status":"ok","timestamp":1787463530580,"user_tz":-330,"elapsed":5886,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"a0281657-cbb4-4da7-abf8-d99a7f8c4931"},"execution_count":44,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["MRI PREDICTION\n"," \n","pituitary : 33.53%\n","notumor : 2.57%\n","meningioma : 2.05%\n","glioma : 61.85%\n"," \n","Prediction : glioma\n","Confidence : 61.85%\n"]},{"output_type":"execute_result","data":{"text/plain":["('glioma', np.float32(61.84885))"]},"metadata":{},"execution_count":44}]},{"cell_type":"code","source":["predict_image(\n"," \"/content/drive/MyDrive/cancer-cnn/extracted data/Testing/meningioma/Te-aug-me_18.jpg\",\n"," model\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":716},"id":"DF6fc_EqfyCU","executionInfo":{"status":"ok","timestamp":1787463530886,"user_tz":-330,"elapsed":305,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"7309bb2c-8019-4bb6-dd8a-9cf31e7a6c02"},"execution_count":45,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["MRI PREDICTION\n"," \n","pituitary : 0.03%\n","notumor : 0.45%\n","meningioma : 1.10%\n","glioma : 98.42%\n"," \n","Prediction : glioma\n","Confidence : 98.42%\n"]},{"output_type":"execute_result","data":{"text/plain":["('glioma', np.float32(98.41791))"]},"metadata":{},"execution_count":45}]},{"cell_type":"code","source":["predict_image(\n"," \"/content/drive/MyDrive/cancer-cnn/extracted data/Testing/notumor/Te-no_127.jpg\",\n"," model\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":716},"id":"qYX-eDWAieBA","executionInfo":{"status":"ok","timestamp":1787463531132,"user_tz":-330,"elapsed":245,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"ad8567c3-c515-40f0-c715-72115e148a46"},"execution_count":46,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["MRI PREDICTION\n"," \n","pituitary : 18.12%\n","notumor : 73.34%\n","meningioma : 7.68%\n","glioma : 0.87%\n"," \n","Prediction : notumor\n","Confidence : 73.34%\n"]},{"output_type":"execute_result","data":{"text/plain":["('notumor', np.float32(73.337296))"]},"metadata":{},"execution_count":46}]},{"cell_type":"code","source":["predict_image(\n"," \"/content/drive/MyDrive/cancer-cnn/extracted data/Testing/pituitary/Te-pi_167.jpg\",\n"," model\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":716},"id":"lUbqlDdqiegI","executionInfo":{"status":"ok","timestamp":1787463531394,"user_tz":-330,"elapsed":264,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"92e1f7a7-e972-4355-aebe-5fe388a400d3"},"execution_count":47,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["MRI PREDICTION\n"," \n","pituitary : 42.87%\n","notumor : 0.08%\n","meningioma : 56.88%\n","glioma : 0.17%\n"," \n","Prediction : meningioma\n","Confidence : 56.88%\n"]},{"output_type":"execute_result","data":{"text/plain":["('meningioma', np.float32(56.880646))"]},"metadata":{},"execution_count":47}]},{"cell_type":"code","source":[],"metadata":{"id":"fSuYzndkjHRk","executionInfo":{"status":"ok","timestamp":1787463531402,"user_tz":-330,"elapsed":8,"user":{"displayName":"Abin","userId":"06944942458087513678"}}},"execution_count":47,"outputs":[]},{"cell_type":"code","source":["print(os.listdir(train_dir))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2m8tygzSqLYM","executionInfo":{"status":"ok","timestamp":1787463531403,"user_tz":-330,"elapsed":8,"user":{"displayName":"Abin","userId":"06944942458087513678"}},"outputId":"dcdf0fc0-b5ce-4c1d-a7b0-9d9074243ce1"},"execution_count":48,"outputs":[{"output_type":"stream","name":"stdout","text":["['pituitary', 'notumor', 'meningioma', 'glioma']\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"j9MzDSuuqS7q"},"execution_count":null,"outputs":[]}],"metadata":{"accelerator":"GPU","colab":{"gpuType":"T4","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"}},"nbformat":4,"nbformat_minor":0} \ No newline at end of file