Project Architecture¶

Enhanced Base Model¶

V3 extends the existing factory pattern architecture with cross-validation capabilities in the BaseModel class.

K-fold Integration

The BaseModel class now includes a cross_validate() method that performs stratified K-fold cross-validation. This method integrates seamlessly with the existing training pipeline while providing robust performance evaluation.

Core Components¶

Enhanced BaseModel Class¶

class BaseModel(ABC):
    def cross_validate(self, X, y, cv_folds=5):
        """Perform stratified K-fold cross-validation"""
        # Creates StratifiedKFold with configured folds
        # Trains model on each fold
        # Returns mean  +-std for all metrics

Configuration¶

Uses existing config.yaml structure with:

cv_folds: 5 for cross-validation configuration
Existing model factory and hyperparameter settings