NMFDivergenceUpdate Class

Class NMFDivergenceUpdate encapulates an NMF update algorithm which minimizes a divergence functional.

Inheritance Hierarchy

SystemObject
CenterSpace.NMath.CoreNMFDivergenceUpdate

Namespace: CenterSpace.NMath.Core
Assembly: NMath (in NMath.dll) Version: 7.4

Syntax

C++

Copy

[SerializableAttribute]
public class NMFDivergenceUpdate : INMFUpdateAlgorithm, 
	ICloneable

<SerializableAttribute>
Public Class NMFDivergenceUpdate
	Implements INMFUpdateAlgorithm, ICloneable

[SerializableAttribute]
public ref class NMFDivergenceUpdate : INMFUpdateAlgorithm, 
	ICloneable

[<SerializableAttribute>]
type NMFDivergenceUpdate = 
    class
        interface INMFUpdateAlgorithm
        interface ICloneable
    end

The NMFDivergenceUpdate type exposes the following members.

Constructors

	Name	Description
	NMFDivergenceUpdate	Constructs a NMFDivergenceUpdate instance.

Top

Methods

	Name	Description
	Clone	Returns a deep copy of self.
	Cost	Returns the value of the cost function for the non-negative factorization of V = WH. The cost function is minimized by the factorization.
	InitialFactors	Provides the initial factors for the iteration process that computes W and H in the non-negative matrix factorization V = WH.
	IterationStep	Updates the factors W and H in the non-negative matrix factorization V = WH.

Top

Remarks

The functional is related to the Poisson likelihood of generating V from W and H:

            D = sum(i,j)Vij*log(Vij/(WH)ij) - Vij + (WH)ij

See the paper Metagenes and Molecular Pattern Discovery Using Matrix Factorization Jean-Philippe Brunet, Pablo Tamayo, Todd R. Golub, and Jill P. Mesirov.

Reference

CenterSpace.NMath.Core Namespace