VotingFeatureSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Ensemble

Voting-based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Instead of trusting just one method to pick features, this approach asks multiple methods to vote. Features that are chosen by many different methods are more likely to be truly important, giving you more confident selections.

How It Works

Combines multiple feature selection methods using voting. Each method casts votes for its top features, and features with the most votes are selected.

Constructors

Constructor Summary
VotingFeatureSelector(int nFeaturesToSelect, int votesPerMethod, int[] columnIndices)

Properties

Property Summary
bool SupportsInverseTransform { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }
int[] VoteCounts { 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)