Click or drag to resize

FactorizationByFastIndependentComponentAnalysis Class

Provides methods for performing Fast Independent Component Analysis (FastICA) and related matrix whitening operations on data matrices.
Inheritance Hierarchy
SystemObject
  Altaxo.Calc.LinearAlgebra.Double.FactorizationFactorizationByFastIndependentComponentAnalysis

Namespace: Altaxo.Calc.LinearAlgebra.Double.Factorization
Assembly: AltaxoCore (in AltaxoCore.dll) Version: 4.8.3572.0 (4.8.3572.0)
Syntax
C#
Copy
public class FactorizationByFastIndependentComponentAnalysis : ILowRankMatrixFactorization, 
	IEquatable<FactorizationByFastIndependentComponentAnalysis>

The FactorizationByFastIndependentComponentAnalysis type exposes the following members.

Constructors
 NameDescription
Public methodFactorizationByFastIndependentComponentAnalysisInitializes a new instance of the FactorizationByFastIndependentComponentAnalysis class
Top
Properties
 NameDescription
Public propertyMaximumNumberOfIterations Gets the maximum number of iterations for the factorization algorithm.
Public propertyTolerance Gets the convergence tolerance.
Top
Methods
 NameDescription
Public methodFactorize Factorizes the input matrix X into a product of two matrices with the specified rank.
Protected methodFinalizeAllows an object to try to free resources and perform other cleanup operations before it is reclaimed by garbage collection.
(Inherited from Object)
Public methodGetTypeGets the Type of the current instance.
(Inherited from Object)
Public methodStatic memberICAWithEvdWhitening Runs a symmetric FastICA implementation based on an eigenvalue-decomposition (EVD) whitening step.
Public methodStatic memberICAWithSvdWhitening Runs a symmetric FastICA implementation using SVD-based whitening with a heuristic choice between full and truncated SVD.
Protected methodMemberwiseCloneCreates a shallow copy of the current Object.
(Inherited from Object)
Public methodStatic memberWhiteningBySvd Performs SVD-based whitening with a heuristic choice between full and truncated SVD:
  • Full SVD when k is large.
  • Truncated SVD when k is significantly smaller than min(n, m).
Top
Remarks
This class includes implementations of the FastICA algorithm for blind source separation, as well as SVD-based whitening utilities.

Reference:

[1] A. Hyvärinen and E. Oja, "A Fast Fixed-Point Algorithm for Independent Component Analysis," in Neural Computation, vol. 9, no. 7, pp. 1483-1492, 10 July 1997, doi:10.1162/neco.1997.9.7.1483.

See Also