VarianceInflationSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Unsupervised

Variance Inflation Factor (VIF) Feature Selection.

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

Implements: IFeatureSelector<T>

For Beginners

VIF measures how much a feature can be predicted by other features. A high VIF means that feature is highly correlated with others (redundant). We remove features with high VIF to keep only the independent, non-redundant ones.

How It Works

Removes features with high multicollinearity by computing the Variance Inflation Factor for each feature and removing those above a threshold.

Constructors

Constructor Summary
VarianceInflationSelector(double vifThreshold, int[] columnIndices)

Properties

Property Summary
Vector<T> VIFScores { get; }
bool SupportsInverseTransform { get; }
double VIFThreshold { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)