Add conditional model training based on configuration

scripts/pipeline.py

@@ -22,22 +22,27 @@ def run_pipeline(config_path: str, log_file: str = None):
         # Run data preprocessing and poisoning
         logger.info("Starting data preprocessing and poisoning")
         pipeline.run()
-
-        # Create and train model
-        logger.info("Creating model")
-        model_factory = ModelFactory()
-        model = model_factory.create_model(config.model_type)
-
-        # Train model
-        logger.info("Training model")
-        X_train, y_train, X_val, y_val = pipeline.get_training_data()
-        model.train(X_train, y_train, X_val, y_val)
-
-        # Save model and metadata
-        model_path = os.path.join(config.model_output, "model")
-        model.save(model_path)
-        config.save(os.path.join(config.model_output, "config.yaml"))
-
+
+        if config.train_model:
+            # Create and train model
+            logger.info("Creating model")
+            model_factory = ModelFactory()
+            model = model_factory.create_model(config.model_type)
+
+            # Train model
+            logger.info("Training model")
+            X_train, y_train, X_val, y_val = pipeline.get_training_data()
+            model.train(X_train, y_train, X_val, y_val)
+
+            # Save model and metadata
+            model_path = os.path.join(config.model_output, "model")
+            model.save(model_path)
+            config.save(os.path.join(config.model_output, "config.yaml"))
+
+            logger.info("Model saved to {}".format(model_path))
+        else:
+            logger.info("Skipping model training")
+
         logger.info("Pipeline completed successfully")
 
     except Exception as e: