PerceptronSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Neural

Perceptron based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

A perceptron is the simplest neural network - just one layer of weights connecting inputs to output. Features with larger weights (positive or negative) are more important for the prediction.

How It Works

Selects features based on their weight magnitudes in a trained single-layer perceptron, identifying features with the strongest learned connections.

Constructors

Constructor Summary
PerceptronSelector(int nFeaturesToSelect, int epochs, double learningRate, int? randomState, int[] columnIndices)

Properties

Property Summary
Vector<T> WeightImportance { get; }
bool SupportsInverseTransform { get; }
double LearningRate { get; }
int Epochs { 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)