edu.ucla.sspace.matrix.factorization

Class SingularValueDecompositionLibJ

java.lang.Object

edu.ucla.sspace.matrix.factorization.AbstractSvd

edu.ucla.sspace.matrix.factorization.SingularValueDecompositionLibJ

All Implemented Interfaces:

MatrixFactorization

public class SingularValueDecompositionLibJ
extends AbstractSvd

A wrapper around the SvdlibjDriver that implements the MatrixFactorization interface.

NOTE: SVDLIBJ is known to have bugs and incorrectly computes the SVD for some dimensions.

Author:

Keith Stevens

Field Summary

Fields inherited from class edu.ucla.sspace.matrix.factorization.AbstractSvd

classFeatures, dataClasses, scaledClassFeatures, scaledDataClasses, singularValues

Constructor Summary

Constructors

Constructor and Description
SingularValueDecompositionLibJ()

Method Summary

Methods

Modifier and Type	Method and Description
void	factorize(MatrixFile mFile, int dimensions) Factorizes the MatrixFile m.
void	factorize(SparseMatrix matrix, int dimensions) Factorizes the SparseMatrix m.
MatrixBuilder	getBuilder() Returns a MatrixBuilder that is optimized for MatrixFiles that will be processed by this MatrixFactorization algorithm.

Methods inherited from class edu.ucla.sspace.matrix.factorization.AbstractSvd

classFeatures, dataClasses, singularValues

Methods inherited from class java.lang.Object

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

Constructor Detail

SingularValueDecompositionLibJ

public SingularValueDecompositionLibJ()

Method Detail

factorize

public void factorize(SparseMatrix matrix,
             int dimensions)

Factorizes the SparseMatrix m. If m is of size MxN, this will generate two matrices: one of size MxnumDimensions and one of size numDimensionsxN. This method must be called before calling dataClasses and classFeatures.

factorize

public void factorize(MatrixFile mFile,
             int dimensions)

Factorizes the MatrixFile m. If m is of size MxN, this will generate two matrices: one of size MxnumDimensions and one of size numDimensionsxN. This method must be called before calling dataClasses and classFeatures.

getBuilder

public MatrixBuilder getBuilder()

Returns a MatrixBuilder that is optimized for MatrixFiles that will be processed by this MatrixFactorization algorithm.