LearningCurveEngine<T>

Models & Types · AiDotNet.Evaluation.Engines

Engine for generating learning curves: how model performance changes with training set size.

For Beginners

Learning curves show how your model improves as you give it more data:

  • X-axis: Number of training samples
  • Y-axis: Performance metric (e.g., accuracy)
  • Typically shows both training and validation scores

How It Works

What learning curves tell you:

  • High bias (underfitting): Both curves plateau at low performance
  • High variance (overfitting): Large gap between train (high) and val (low)
  • Need more data: Validation curve still improving at max training size
  • Good fit: Both curves converge at high performance

Constructors

Constructor Summary
LearningCurveEngine(LearningCurveOptions options) Initializes the learning curve engine.

Methods

Method Summary
LearningCurveResult<T> Generate(T[] features, T[] targets, Func<T[], T[], TModel> trainFunc, Func<TModel, T[], T[]> predictFunc, string metricName, double[] trainSizes, int cvFolds, bool isClassification, bool higherIsBetter) Generates a learning curve by training at various dataset sizes.