ICABasedSelector<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Projection
Independent Component Analysis (ICA) based Feature Selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → ICABasedSelector<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
ICA finds hidden “sources” that combine to create your observed data (like separating mixed audio signals). Features that strongly contribute to these independent sources capture unique, non-redundant information and are good candidates for selection.
How It Works
Uses ICA to find independent components and selects features that contribute most to these statistically independent sources.
Constructors
| Constructor | Summary |
|---|---|
ICABasedSelector(int nFeaturesToSelect, int nComponents, int nIterations, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> FeatureScores { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |