model_trainer.py

import joblib
from sklearn.ensemble import IsolationForest
import os

def train_isolation_forest(df, sensor_cols, contamination=0.1):
    """
    Train Isolation Forest model for anomaly detection
    """
    print("Training Isolation Forest model...")

    # Initialize and train the model
    iso_forest = IsolationForest(
        contamination=contamination,
        random_state=42,
        n_estimators=100
    )

    iso_forest.fit(df[sensor_cols])

    # Predict anomalies and scores
    df['anomaly'] = iso_forest.predict(df[sensor_cols])
    df['anomaly_score'] = iso_forest.decision_function(df[sensor_cols])

    print("Model training completed!")
    return iso_forest, df

def save_model(model, filepath='isolation_forest_model.pkl'):
    """
    Save trained model to disk
    """
    joblib.dump(model, filepath)
    print(f"Model saved to {filepath}")

def load_model(filepath='isolation_forest_model.pkl'):
    """
    Load trained model from disk
    """
    if not os.path.exists(filepath):
        return None

    model = joblib.load(filepath)
    print(f"Model loaded from {filepath}")
    return model

def add_anomaly_scores(df, model, sensor_cols):
    """
    Add anomaly predictions to dataframe
    """
    df['anomaly'] = model.predict(df[sensor_cols])
    df['anomaly_score'] = model.decision_function(df[sensor_cols])
    return df

if __name__ == "__main__":
    # Test the model trainer
    try:
        from data_processor import load_and_process_data
        df, sensor_cols = load_and_process_data()
        model, df_with_anomalies = train_isolation_forest(df, sensor_cols)
        save_model(model)
        print("Model training and saving completed successfully!")
    except Exception as e:
        print(f"Error in model training: {e}")