ExponentialDecaySelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Streaming

Exponential Decay Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Unlike a sharp sliding window, exponential decay smoothly reduces the importance of older samples. Recent data has the most influence, but older data still contributes a little. This creates a smooth transition as the data distribution changes.

How It Works

Uses exponentially weighted statistics for feature selection, giving more importance to recent samples while gradually forgetting older ones.

Constructors

Constructor Summary
ExponentialDecaySelector(int nFeaturesToSelect, double decayFactor, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureScores { get; }
bool SupportsInverseTransform { get; }
double DecayFactor { 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)
void PartialFit(Matrix<T> data, Vector<T> target)