MultiSURF<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Filter.Relief

MultiSURF algorithm with adaptive distance thresholds per instance.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ†’ MultiSURF<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

While SURF uses one global distance threshold, MultiSURF calculates a custom threshold for each sample based on how spread out its neighbors are. This works better for datasets with varying density regions.

How It Works

MultiSURF improves SURF by using instance-specific distance thresholds based on the local density around each instance. This adapts to varying data densities across the feature space.

Constructors

Constructor Summary
MultiSURF(int nFeaturesToSelect, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureWeights { get; }
bool SupportsInverseTransform { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Matrix<T> data, Vector<T> target)