{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":14774,"databundleVersionId":875431,"sourceType":"competition"}],"dockerImageVersionId":30887,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport numpy as np\nimport pandas as pd\nimport tensorflow as tf\nimport matplotlib.pyplot as plt\nfrom sklearn.metrics import classification_report, roc_curve, auc\n\nImageDataGenerator = tf.keras.preprocessing.image.ImageDataGenerator\nResNet152V2 = tf.keras.applications.ResNet152V2\nDense = tf.keras.layers.Dense\nGlobalAveragePooling2D = tf.keras.layers.GlobalAveragePooling2D\nModel = tf.keras.models.Model\nAdam = tf.keras.optimizers.Adam\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import f1_score, roc_auc_score","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T09:09:44.568892Z","iopub.execute_input":"2025-03-03T09:09:44.569257Z","iopub.status.idle":"2025-03-03T09:09:44.573856Z","shell.execute_reply.started":"2025-03-03T09:09:44.569226Z","shell.execute_reply":"2025-03-03T09:09:44.573137Z"}},"outputs":[],"execution_count":6},{"cell_type":"code","source":"# Step 1: Load and Prepare the Data\nbase_dir = \"/kaggle/input/aptos2019-blindness-detection\"\ntrain_images_dir = os.path.join(base_dir, \"train_images\")\ntrain_csv_path = os.path.join(base_dir, \"train.csv\")\n\ndf = pd.read_csv(train_csv_path)\ndf[\"id_code\"] = df[\"id_code\"].apply(lambda x: os.path.join(train_images_dir, f\"{x}.png\"))\ndf[\"diagnosis\"] = df[\"diagnosis\"].astype(str)  # Ensure labels are string type\n\n# Step 2: Split the Data\ntrain_df, val_df = train_test_split(df, test_size=0.2, random_state=42, stratify=df[\"diagnosis\"])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T09:09:56.289044Z","iopub.execute_input":"2025-03-03T09:09:56.289546Z","iopub.status.idle":"2025-03-03T09:09:56.326711Z","shell.execute_reply.started":"2025-03-03T09:09:56.289515Z","shell.execute_reply":"2025-03-03T09:09:56.325816Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"# Step 3: Data Generators\ndatagen = ImageDataGenerator(rescale=1.0 / 255)\n\ntrain_generator = datagen.flow_from_dataframe(\n    dataframe=train_df,\n    x_col=\"id_code\",\n    y_col=\"diagnosis\",\n    target_size=(224, 224),\n    batch_size=32,\n    class_mode=\"categorical\",\n    shuffle=True\n)\n\nval_generator = datagen.flow_from_dataframe(\n    dataframe=val_df,\n    x_col=\"id_code\",\n    y_col=\"diagnosis\",\n    target_size=(224, 224),\n    batch_size=32,\n    class_mode=\"categorical\",\n    shuffle=False  # Important for correct evaluation\n)\n\n# Get number of classes dynamically\nnum_classes = len(train_generator.class_indices)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T09:10:08.229612Z","iopub.execute_input":"2025-03-03T09:10:08.229895Z","iopub.status.idle":"2025-03-03T09:10:15.032796Z","shell.execute_reply.started":"2025-03-03T09:10:08.229872Z","shell.execute_reply":"2025-03-03T09:10:15.031887Z"}},"outputs":[{"name":"stdout","text":"Found 2929 validated image filenames belonging to 5 classes.\nFound 733 validated image filenames belonging to 5 classes.\n","output_type":"stream"}],"execution_count":8},{"cell_type":"code","source":"# Step 4: Define Model Architecture\nbase_model = ResNet152V2(weights=\"imagenet\", include_top=False, input_shape=(224, 224, 3))\nx = base_model.output\nx = GlobalAveragePooling2D()(x)\nx = Dense(256, activation=\"relu\")(x)\npredictions = Dense(num_classes, activation=\"softmax\")(x)\n\nmodel = Model(inputs=base_model.input, outputs=predictions)\n\n# Freeze base_model layers\nfor layer in base_model.layers:\n    layer.trainable = False\n\n# Compile the model\nmodel.compile(optimizer=Adam(learning_rate=0.001), loss=\"categorical_crossentropy\", metrics=[\"accuracy\"])\n\n# Step 5: Train the Model\nhistory = model.fit(\n    train_generator,\n    validation_data=val_generator,\n    epochs=10\n)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T09:10:20.080727Z","iopub.execute_input":"2025-03-03T09:10:20.081015Z","iopub.status.idle":"2025-03-03T10:08:29.833728Z","shell.execute_reply.started":"2025-03-03T09:10:20.080992Z","shell.execute_reply":"2025-03-03T10:08:29.832879Z"}},"outputs":[{"name":"stdout","text":"Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet152v2_weights_tf_dim_ordering_tf_kernels_notop.h5\n\u001b[1m234545216/234545216\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\nEpoch 1/10\n","output_type":"stream"},{"name":"stderr","text":"/usr/local/lib/python3.10/dist-packages/keras/src/trainers/data_adapters/py_dataset_adapter.py:122: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n  self._warn_if_super_not_called()\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m454s\u001b[0m 4s/step - accuracy: 0.6505 - loss: 1.0185 - val_accuracy: 0.7312 - val_loss: 0.7386\nEpoch 2/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m338s\u001b[0m 3s/step - accuracy: 0.7917 - loss: 0.5554 - val_accuracy: 0.7517 - val_loss: 0.6644\nEpoch 3/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m336s\u001b[0m 3s/step - accuracy: 0.8102 - loss: 0.5148 - val_accuracy: 0.7613 - val_loss: 0.6929\nEpoch 4/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m336s\u001b[0m 3s/step - accuracy: 0.8493 - loss: 0.4064 - val_accuracy: 0.7531 - val_loss: 0.6793\nEpoch 5/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m336s\u001b[0m 3s/step - accuracy: 0.8518 - loss: 0.4027 - val_accuracy: 0.7299 - val_loss: 0.7203\nEpoch 6/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m335s\u001b[0m 3s/step - accuracy: 0.8630 - loss: 0.3616 - val_accuracy: 0.7626 - val_loss: 0.6835\nEpoch 7/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m335s\u001b[0m 3s/step - accuracy: 0.9049 - loss: 0.2750 - val_accuracy: 0.7258 - val_loss: 0.7454\nEpoch 8/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m335s\u001b[0m 3s/step - accuracy: 0.9000 - loss: 0.2792 - val_accuracy: 0.7708 - val_loss: 0.7604\nEpoch 9/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m338s\u001b[0m 3s/step - accuracy: 0.9211 - loss: 0.2295 - val_accuracy: 0.7599 - val_loss: 0.7857\nEpoch 10/10\n\u001b[1m92/92\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m335s\u001b[0m 3s/step - accuracy: 0.9070 - loss: 0.2373 - val_accuracy: 0.7503 - val_loss: 0.8652\n","output_type":"stream"}],"execution_count":9},{"cell_type":"code","source":"# Step 6: Save the Model\nmodel.save(\"blindness_detection_model.h5\")\n\n# Step 7: Evaluate the Model\nval_loss, val_accuracy = model.evaluate(val_generator)\nprint(f\"Validation Loss: {val_loss}\")\nprint(f\"Validation Accuracy: {val_accuracy}\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T10:08:29.834934Z","iopub.execute_input":"2025-03-03T10:08:29.835244Z","iopub.status.idle":"2025-03-03T10:09:43.634844Z","shell.execute_reply.started":"2025-03-03T10:08:29.835214Z","shell.execute_reply":"2025-03-03T10:09:43.633933Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m67s\u001b[0m 3s/step - accuracy: 0.7357 - loss: 0.9102\nValidation Loss: 0.8651968240737915\nValidation Accuracy: 0.7503410577774048\n","output_type":"stream"}],"execution_count":10},{"cell_type":"code","source":"from sklearn.metrics import f1_score\n\n# Get true labels\ny_true = np.array(val_generator.classes)  # Ensure it's a NumPy array\n\n# Get model predictions\ny_pred_proba = model.predict(val_generator)  # Get predicted probabilities\ny_pred = np.argmax(y_pred_proba, axis=1)  # Convert probabilities to class labels\n\n# Compute F1-score\nf1 = f1_score(y_true, y_pred, average=\"weighted\")\nprint(f\"F1 Score: {f1:.4f}\")\n\n# Compute ROC-AUC for each class\nroc_auc_values = []\nplt.figure(figsize=(8, 6))\n\nfor i in range(num_classes):\n    y_true_bin = (y_true == i).astype(int)  # Convert to binary labels\n    fpr, tpr, _ = roc_curve(y_true_bin, y_pred_proba[:, i])\n    roc_auc = auc(fpr, tpr)\n    roc_auc_values.append(roc_auc)\n\n    # Plot ROC curve for the class\n    plt.plot(fpr, tpr, label=f\"Class {i} (AUC = {roc_auc:.2f})\")\n\n# Plot reference line\nplt.plot([0, 1], [0, 1], 'k--')\n\n# Labels and legend\nplt.xlabel(\"False Positive Rate\")\nplt.ylabel(\"True Positive Rate\")\nplt.title(\"ROC-AUC Curve\")\nplt.legend()\nplt.show()\n\n# Print ROC-AUC scores for each class\nfor i, auc_value in enumerate(roc_auc_values):\n    print(f\"Class {i} - ROC AUC Score: {auc_value:.4f}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-03-03T10:42:11.214027Z","iopub.execute_input":"2025-03-03T10:42:11.2144Z","iopub.status.idle":"2025-03-03T10:43:24.976618Z","shell.execute_reply.started":"2025-03-03T10:42:11.214371Z","shell.execute_reply":"2025-03-03T10:43:24.975845Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m67s\u001b[0m 3s/step\nF1 Score: 0.7280\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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\n"},"metadata":{}},{"name":"stdout","text":"Class 0 - ROC AUC Score: 0.9843\nClass 1 - ROC AUC Score: 0.8587\nClass 2 - ROC AUC Score: 0.9020\nClass 3 - ROC AUC Score: 0.8844\nClass 4 - ROC AUC Score: 0.8691\n","output_type":"stream"}],"execution_count":13}]}