Index

A C D E G L M N O R S U
All Classes All Packages

A

AlternatingLeastSquaresMatrixFactorization - Class in org.dulab.javanmf.algorithms

This class performs non-negative matrix factorization using the alternating non-negative least squares method.

AlternatingLeastSquaresMatrixFactorization(double, int) - Constructor for class org.dulab.javanmf.algorithms.AlternatingLeastSquaresMatrixFactorization

Creates an instance of AlternatingLeastSquaresMatrixFactorization

AlternatingLeastSquaresMatrixFactorization(Constraint, Constraint, double, int) - Constructor for class org.dulab.javanmf.algorithms.AlternatingLeastSquaresMatrixFactorization

Creates an instance of AlternatingLeastSquaresMatrixFactorization

apply(DMatrixRMaj) - Method in interface org.dulab.javanmf.algorithms.Constraint

apply(DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.DefaultConstraint

C

Constraint - Interface in org.dulab.javanmf.algorithms

D

decompose(DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.SingularValueDecomposition

Performs non-negative singular value decomposition (NNDSVD) of matrix X.

DefaultConstraint - Class in org.dulab.javanmf.algorithms

DefaultConstraint() - Constructor for class org.dulab.javanmf.algorithms.DefaultConstraint

E

EuclideanDistance - Class in org.dulab.javanmf.measures

Calculates the distance between matrices X and WH using the euclidean distance || X − WH ||²

EuclideanDistance() - Constructor for class org.dulab.javanmf.measures.EuclideanDistance

execute(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.MatrixFactorization

Performs the non-negative matrix factorization with given initial matrices W and H.

execute(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, boolean) - Method in class org.dulab.javanmf.algorithms.MatrixFactorization

Performs the non-negative matrix factorization with given initial matrices W and H.

G

get(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.measures.EuclideanDistance

get(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.measures.Measure

Returns distance between matrices X and WH

L

lambda - Variable in class org.dulab.javanmf.updaterules.RegularizationUpdateRule

l₁-regularization coefficient

M

MatrixFactorization - Class in org.dulab.javanmf.algorithms

This class performs non-negative matrix factorization: for given matrix X, find matrices W and H that minimize the objective function

MatrixFactorization(UpdateRule, UpdateRule, double, int) - Constructor for class org.dulab.javanmf.algorithms.MatrixFactorization

Creates an instance of MatrixFactorization

MatrixRegression - Class in org.dulab.javanmf.algorithms

This class performs non-negative matrix regression: for given matrices X and W, find matrix H that minimizes the objective function

MatrixRegression(UpdateRule, double, int) - Constructor for class org.dulab.javanmf.algorithms.MatrixRegression

Creates an instance of MatrixRegression

MatrixUtils - Class in org.dulab.javanmf.algorithms

MatrixUtils() - Constructor for class org.dulab.javanmf.algorithms.MatrixUtils

measure - Variable in class org.dulab.javanmf.updaterules.UpdateRule

Instance of Measure associated with this update rule

Measure - Class in org.dulab.javanmf.measures

Provides a template for calculating the distance between matrices X and WH

Measure() - Constructor for class org.dulab.javanmf.measures.Measure

minimumEquals(DMatrixRMaj, DMatrixRMaj) - Static method in class org.dulab.javanmf.algorithms.MatrixUtils

mu - Variable in class org.dulab.javanmf.updaterules.RegularizationUpdateRule

l₂-regularization coefficient

MUpdateRule - Class in org.dulab.javanmf.updaterules

Performs multiplicative update for the euclidean distance with regularization

MUpdateRule(double, double) - Constructor for class org.dulab.javanmf.updaterules.MUpdateRule

Creates an instance of MUpdateRule with given regularization coefficients

N

NonNegativeLeastSquares - Class in org.dulab.javanmf.algorithms

This class solves the non-negative least squares problem using the active set method.

NonNegativeLeastSquares() - Constructor for class org.dulab.javanmf.algorithms.NonNegativeLeastSquares

O

org.dulab.javanmf.algorithms - package org.dulab.javanmf.algorithms

Provides classes for performing non-negative matrix factorization, non-negative matrix regression, and non-negative singular value decomposition.

org.dulab.javanmf.measures - package org.dulab.javanmf.measures

Provides classes for estimating the distance between matrices X and WH.

org.dulab.javanmf.updaterules - package org.dulab.javanmf.updaterules

Provides classes for updating matrix H in the direction of minimising the distance D(X, WH).

R

RegularizationUpdateRule - Class in org.dulab.javanmf.updaterules

Provides a template for updating matrix H in the direction of minimizing the distance D(X, WH) with regularization

RegularizationUpdateRule(Measure, double, double) - Constructor for class org.dulab.javanmf.updaterules.RegularizationUpdateRule

Creates an instance of RegularizationUpdateRule with given regularization coefficients

S

SingularValueDecomposition - Class in org.dulab.javanmf.algorithms

This class performs non-negative singular value decomposition: first, the singular value decomposition is performed; then, the non-negative matrices W and H are formed.

SingularValueDecomposition(DMatrixRMaj) - Constructor for class org.dulab.javanmf.algorithms.SingularValueDecomposition

Creates an instance of SingularValueDecomposition for given matrix

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.AlternatingLeastSquaresMatrixFactorization

Performs non-negative matrix regression with the upper limit constraint

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.MatrixRegression

Performs non-negative matrix regression

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.NonNegativeLeastSquares

Finds matrix D such that D = argmin || X - Z x D ||^2

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, boolean) - Method in class org.dulab.javanmf.algorithms.AlternatingLeastSquaresMatrixFactorization

Performs non-negative matrix regression with the upper limit constraint

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, boolean) - Method in class org.dulab.javanmf.algorithms.MatrixRegression

Performs non-negative matrix regression

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.algorithms.MatrixRegression

Performs non-negative matrix regression with the upper limit constraint

solve(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, DMatrixRMaj, boolean) - Method in class org.dulab.javanmf.algorithms.MatrixRegression

Performs non-negative matrix regression with the upper limit constraint

U

update(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.updaterules.MUpdateRule

update(DMatrixRMaj, DMatrixRMaj, DMatrixRMaj) - Method in class org.dulab.javanmf.updaterules.UpdateRule

Updates matrix H to minimize distance between X and WH

UpdateRule - Class in org.dulab.javanmf.updaterules

Provides a template for updating matrix H in the direction of minimizing the distance D(X, WH)

UpdateRule(Measure) - Constructor for class org.dulab.javanmf.updaterules.UpdateRule

Creates an instance of UpdateRule

A C D E G L M N O R S U
All Classes All Packages