HuberCorrelationSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Robust

Huber Correlation-based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Regular correlation can be strongly affected by outliers. The Huber method treats moderate deviations normally but reduces the influence of extreme values. This gives you feature-target correlations that better reflect the typical relationship, not just outlier effects.

How It Works

Uses Huber M-estimator for robust correlation computation, downweighting the influence of outliers in the feature-target relationship.

Constructors

Constructor Summary
HuberCorrelationSelector(int nFeaturesToSelect, double huberDelta, int[] columnIndices)

Properties

Property Summary
Vector<T> RobustCorrelations { get; }
bool SupportsInverseTransform { get; }
double HuberDelta { 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)