Validation

In more detail

Validation data is not used for training. It helps monitor performance during training (validation set) or final evaluation (test set). Common techniques: hold-out validation, k-fold cross-validation, time series cross-validation. Essential for preventing overfitting and selecting best models.

Examples

• Train-validation-test split
• K-fold cross-validation
• Time series cross-validation
• Stratified k-fold for imbalanced data

Related terms