DataSplitterBase<T>

Base Classes · AiDotNet.Preprocessing.DataPreparation

Base class for all data splitters providing common functionality.

Implements: IDataSplitter<T>

Known derived types: AdversarialValidationSplitter<T>, AnchoredWalkForwardSplitter<T>, AntiClusteringSplitter<T>, BalancedSplitter<T>, BlockedTimeSeriesSplitter<T>, Bootstrap632PlusSplitter<T>, BootstrapSplitter<T>, ClusterBasedSplitter<T>, CombinatorialPurgedSplitter<T>, CovariateShiftSplitter<T>, DirichletSplitter<T>, DistributionPreservingSplitter<T>, DoubleCVSplitter<T>, GraphSplitter<T>, GroupKFoldSplitter<T>, GroupShuffleSplitter<T>, HierarchicalSplitter<T>, HoldoutSplitter<T>, IIDClientSplitter<T>, ImagePatchSplitter<T>, IterativeStratificationSplitter<T>, KFoldSplitter<T>, LandmarkWindowSplitter<T>, LeaveOneGroupOutSplitter<T>, LeaveOneOutSplitter<T>, LeavePGroupsOutSplitter<T>, LeavePOutSplitter<T>, MultiTaskSplitter<T>, NestedCVSplitter<T>, NonIIDClientSplitter<T>, OnlineSlidingWindowSplitter<T>, OutOfDistributionSplitter<T>, PoolBasedSplitter<T>, PredefinedSplitter<T>, PrequentialSplitter<T>, PurgedKFoldSplitter<T>, PurgedWalkForwardSplitter<T>, QueryByCommitteeSplitter<T>, RepeatedKFoldSplitter<T>, RollingOriginSplitter<T>, SequenceSplitter<T>, ShuffleSplitter<T>, SimilarityBasedSplitter<T>, SlidingWindowSplitter<T>, SpatialSplitter<T>, StratifiedBootstrapSplitter<T>, StratifiedGroupKFoldSplitter<T>, StratifiedKFoldSplitter<T>, StratifiedRepeatedKFoldSplitter<T>, StratifiedShuffleSplitter<T>, StratifiedTrainTestSplitter<T>, StratifiedTrainValTestSplitter<T>, TemporalSpatialSplitter<T>, TimeSeriesSplitter<T>, TrainTestSplitter<T>, TrainValTestSplitter<T>, WalkForwardSplitter<T>

For Beginners

This base class provides shared utilities that all data splitters need:

  • Shuffling indices randomly
  • Selecting rows from matrices
  • Validating inputs
  • Working with both Matrix and Tensor data

How It Works

When creating a custom splitter, inherit from this class to get these utilities for free. You only need to implement the specific splitting logic for your algorithm.

Properties

Property Summary
bool RequiresLabels { get; }
bool SupportsValidation { get; }
int NumSplits { get; }
string Description { get; }

Methods

Method Summary
DataSplitResult<T> Split(Matrix<T> X, Vector<T> y)
IEnumerable<DataSplitResult<T>> GetSplits(Matrix<T> X, Vector<T> y)
IEnumerable<TensorSplitResult<T>> GetTensorSplits(Tensor<T> X, Tensor<T> y)
TensorSplitResult<T> SplitTensor(Tensor<T> X, Tensor<T> y)