baseline_model.py

"""
Baseline Model Module
Trains and evaluates a baseline Random Forest model.
"""

import numpy as np
import time
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report

from data_loader import DataLoader


class BaselineModel:
    """Creates and evaluates a baseline Random Forest classifier."""
    
    def __init__(self):
        """Initialize the baseline model."""
        self.model = None
        self.train_time = None
        self.accuracy = None
    
    def create_baseline_model(self):
        """Create a Random Forest classifier with default parameters."""
        self.model = RandomForestClassifier(n_estimators=50, max_depth=3, random_state=42)
        return self.model
    
    def train(self, X_train, y_train):
        """Train the baseline model and record training time."""
        if self.model is None:
            self.create_baseline_model()
        
        print("Training baseline model...")
        start_time = time.time()
        self.model.fit(X_train, y_train)
        self.train_time = time.time() - start_time
        print(f"Training completed in {self.train_time:.2f} seconds")
    
    def evaluate(self, X_test, y_test):
        """Evaluate the model on test data."""
        predictions = self.model.predict(X_test)
        self.accuracy = accuracy_score(y_test, predictions)
        
        print("\n" + "=" * 50)
        print("Baseline Model Evaluation")
        print("=" * 50)
        print(f"Accuracy: {self.accuracy:.4f} ({self.accuracy*100:.2f}%)")
        print(f"Training Time: {self.train_time:.2f} seconds")
        print("\nClassification Report:")
        print(classification_report(y_test, predictions, 
                                   target_names=['No Churn', 'Churn']))
        print("=" * 50)
        
        return self.accuracy
    
    def save_results(self):
        """Save baseline results for later comparison."""
        results = {
            'accuracy': self.accuracy,
            'train_time': self.train_time,
            'model_params': self.model.get_params()
        }
        np.save('output/baseline_results.npy', results)
        print("Baseline results saved to output/baseline_results.npy")


if __name__ == "__main__":
    import os
    os.makedirs('output', exist_ok=True)
    
    loader = DataLoader()
    X_train, X_test, y_train, y_test = loader.load_data()
    loader.get_data_info()
    
    baseline = BaselineModel()
    baseline.create_baseline_model()
    baseline.train(X_train, y_train)
    baseline.evaluate(X_test, y_test)
    baseline.save_results()