Package org.dulab.javanmf.algorithms

Class AlternatingLeastSquaresMatrixFactorization

java.lang.Object

org.dulab.javanmf.algorithms.AlternatingLeastSquaresMatrixFactorization

public class AlternatingLeastSquaresMatrixFactorization
extends java.lang.Object

This class performs non-negative matrix factorization using the alternating non-negative least squares method. See H. Kim and H. Park "NON-NEGATIVE MATRIX FACTORIZATION BASED ON ALTERNATING NON-NEGATIVITY CONSTRAINED LEAST SQUARES AND ACTIVE SET METHOD"

Author:

Du-Lab Team dulab.binf@gmail.com

Constructor Summary

Constructors

Constructor	Description
AlternatingLeastSquaresMatrixFactorization(double tolerance, int maxIteration)	Creates an instance of AlternatingLeastSquaresMatrixFactorization
AlternatingLeastSquaresMatrixFactorization(Constraint wtConstraint, Constraint hConstraint, double tolerance, int maxIteration)	Creates an instance of AlternatingLeastSquaresMatrixFactorization

Method Summary

Modifier and Type	Method	Description
void	solve(org.ejml.data.DMatrixRMaj x, org.ejml.data.DMatrixRMaj w, org.ejml.data.DMatrixRMaj h)	Performs non-negative matrix regression with the upper limit constraint
void	solve(org.ejml.data.DMatrixRMaj x, org.ejml.data.DMatrixRMaj w, org.ejml.data.DMatrixRMaj h, boolean verbose)	Performs non-negative matrix regression with the upper limit constraint

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

AlternatingLeastSquaresMatrixFactorization

public AlternatingLeastSquaresMatrixFactorization(@Nullable Constraint wtConstraint, @Nullable Constraint hConstraint, double tolerance, int maxIteration)

Creates an instance of AlternatingLeastSquaresMatrixFactorization

Parameters:

wtConstraint - instance of Constraint for matrix W^T

hConstraint - instance of Constraint for matrix H

tolerance - the fitting error tolerance

maxIteration - maximum number of iterations to use

AlternatingLeastSquaresMatrixFactorization

public AlternatingLeastSquaresMatrixFactorization(double tolerance, int maxIteration)

Creates an instance of AlternatingLeastSquaresMatrixFactorization

Parameters:

tolerance - the fitting error tolerance

maxIteration - maximum number of iterations to use

Method Details

solve

public void solve(@Nonnull org.ejml.data.DMatrixRMaj x, @Nonnull org.ejml.data.DMatrixRMaj w, @Nonnull org.ejml.data.DMatrixRMaj h, boolean verbose)

Performs non-negative matrix regression with the upper limit constraint

Parameters:

x - matrix of shape [N_points, N_vectors], a collection of vectors in N_points-dimensional space

w - matrix of shape [N_points, N_components], a collection of components

h - matrix of shape [N_components, N_vectors], a collection of the decomposition coefficients

verbose - flag to output verbose information

solve

public void solve(@Nonnull org.ejml.data.DMatrixRMaj x, @Nonnull org.ejml.data.DMatrixRMaj w, @Nonnull org.ejml.data.DMatrixRMaj h)

Performs non-negative matrix regression with the upper limit constraint

Parameters:

x - matrix of shape [N_points, N_vectors], a collection of vectors in N_points-dimensional space

w - matrix of shape [N_points, N_components], a collection of components

h - matrix of shape [N_components, N_vectors], a collection of the decomposition coefficients