ValidationCurveEngine<T>

Models & Types · AiDotNet.Evaluation.Engines

Engine for generating validation curves: how model performance changes with hyperparameter values.

For Beginners

Validation curves show how a hyperparameter affects performance:

  • X-axis: Hyperparameter values (e.g., regularization strength)
  • Y-axis: Performance metric (training and validation scores)
  • Helps identify optimal hyperparameter range

How It Works

What validation curves tell you:

  • Underfitting region: Both train and val scores are low
  • Optimal region: Val score peaks, train is slightly higher
  • Overfitting region: Train high, val declining

Constructors

Constructor Summary
ValidationCurveEngine(ValidationCurveOptions options) Initializes the validation curve engine.

Methods

Method Summary
ValidationCurveResult<T> Generate(T[] features, T[] targets, Func<T[], T[], double, TModel> trainFunc, Func<TModel, T[], T[]> predictFunc, string parameterName, double[] parameterValues, string metricName, int cvFolds, bool isClassification, bool higherIsBetter) Generates a validation curve for a given hyperparameter.