MonteCarloStrategy<T>

Models & Types · AiDotNet.Evaluation.CrossValidation

Monte Carlo cross-validation (repeated random sub-sampling validation).

Implements: ICrossValidationStrategy<T>

For Beginners

Monte Carlo CV randomly splits data multiple times:

  • Each iteration creates a random train/test split
  • Unlike K-Fold, samples may appear in validation multiple times or never
  • More flexible control over train/test sizes
  • Good for small datasets where K-Fold has high variance

How It Works

Comparison to K-Fold:

  • K-Fold: Each sample in validation exactly once
  • Monte Carlo: Samples may repeat or be missing in validation
  • Monte Carlo: More iterations possible with less computation

Constructors

Constructor Summary
MonteCarloStrategy(int nIterations, double testSize, int? randomSeed) Initializes Monte Carlo cross-validation.

Properties

Property Summary
int NumSplits { get; }
string Description { get; }
string Name { get; }

Methods

Method Summary
IEnumerable<ValueTuple<int[], int[]>> Split(int dataSize, ReadOnlySpan<T> labels)