LLEBasedSelector<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Manifold
Locally Linear Embedding (LLE) based Feature Selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → LLEBasedSelector<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
LLE assumes data lies on a curved surface (manifold) in high-dimensional space. It finds how to reconstruct each point from its neighbors. Features that are most important for these local reconstructions are selected.
How It Works
Uses Locally Linear Embedding to identify features that best preserve the local neighborhood structure of the data manifold.
Constructors
| Constructor | Summary |
|---|---|
LLEBasedSelector(int nFeaturesToSelect, int nNeighbors, int[] columnIndices, double regularization) |
Properties
| Property | Summary |
|---|---|
Vector<T> FeatureScores { get; } |
|
bool SupportsInverseTransform { get; } |
|
double Regularization { get; } |
Gets the regularization of the local Gram matrices, relative to their trace (scikit-learn’s reg). |
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |