HierarchicalFS<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Clustering

Hierarchical Clustering-based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

This method builds a family tree of features, starting with each feature as its own group, then gradually merging the most similar ones. When we cut this tree at a certain height, we get groups of related features. We pick the best feature from each group.

How It Works

Uses hierarchical (agglomerative) clustering to build a tree of feature relationships, then cuts the tree to create groups and selects representative features from each group.

Constructors

Constructor Summary
HierarchicalFS(int nFeaturesToSelect, string linkage, int[] columnIndices)

Properties

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

Methods

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