AttentionBasedSelector<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Neural

Attention-based Feature Selection using neural attention mechanisms.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Attention is like a spotlight that tells the model where to focus. Features that get more “spotlight” (higher attention) are more important for making predictions. This method learns those importance weights automatically from the data.

How It Works

Uses an attention mechanism to learn feature importance weights. The attention scores indicate which features the model focuses on when making predictions.

Constructors

Constructor Summary
AttentionBasedSelector(int nFeaturesToSelect, int hiddenSize, int nEpochs, double learningRate, int[] columnIndices)

Properties

Property Summary
Vector<T> AttentionScores { get; }
bool SupportsInverseTransform { 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)