TruncatedSVDSelection<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Dimensionality

Truncated SVD Feature Selection using singular value decomposition.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

SVD breaks down your data into components that explain different directions of variation. By looking at which original features contribute to the most important directions, we can select features that capture the most information about your data.

How It Works

Uses Singular Value Decomposition to identify which original features contribute most to the principal directions of variation. Unlike PCA, this works directly on the data matrix without centering.

Constructors

Constructor Summary
TruncatedSVDSelection(int nFeaturesToSelect, int nComponents, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureImportances { get; }
bool SupportsInverseTransform { get; }
int NComponents { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

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