EntropyBasedSelection<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter

Entropy-Based Feature Selection using discretized feature entropy.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → EntropyBasedSelection<T>

Implements: IFixedCountFeatureSelector<T>

For Beginners

Entropy measures how “mixed up” or unpredictable a feature’s values are. A feature that’s always the same value has zero entropy and tells you nothing useful. Features with higher entropy have more variety and can better distinguish between different samples.

How It Works

Measures the information content of each feature using Shannon entropy. Features with higher entropy contain more information and discriminative power. Features with very low entropy (constant or near-constant) are removed.

Constructors

Constructor Summary
EntropyBasedSelection(int nFeaturesToSelect, int nBins, double minEntropyRatio, int[] columnIndices)

Properties

Property Summary
bool SupportsInverseTransform { get; }
double[] EntropyScores { get; }
int NBins { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

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