no.uib.cipr.matrix
Class DenseLU

java.lang.Object
  no.uib.cipr.matrix.DenseLU

public class DenseLU
extends java.lang.Object

Dense LU decomposition FIXME: DenseLU is broken! Fix it

Constructor Summary

DenseLU(int m, int n)
Constructor for DenseLU

Method Summary

DenseLU	factor(DenseMatrix A) Creates an LU decomposition of the given matrix
static DenseLU	factorize(Matrix A) Creates an LU decomposition of the given matrix
UnitLowerTriangDenseMatrix	getL() Returns the lower triangular factor
DenseMatrix	getLU() Returns the decomposition matrix
int[]	getPivots() Returns the row pivots
UpperTriangDenseMatrix	getU() Returns the upper triangular factor
boolean	isSingular() Checks for singularity
double	rcond(Matrix A, Matrix.Norm norm) Computes the reciprocal condition number, using either the infinity norm of the 1 norm.
DenseMatrix	solve(DenseMatrix B) Computes A\B, overwriting B
DenseMatrix	transSolve(DenseMatrix B) Computes AT\B, overwriting B

Methods inherited from class java.lang.Object

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

Constructor Detail

DenseLU

public DenseLU(int m,
               int n)

Constructor for DenseLU

Parameters:

m - Number of rows

n - Number of columns

Method Detail

factorize

public static DenseLU factorize(Matrix A)

Creates an LU decomposition of the given matrix

Parameters:

A - Matrix to decompose. Not modified

Returns:

The current decomposition

factor

public DenseLU factor(DenseMatrix A)

Creates an LU decomposition of the given matrix

Parameters:

A - Matrix to decompose. Overwritten with the decomposition

Returns:

The current decomposition

getL

public UnitLowerTriangDenseMatrix getL()

Returns the lower triangular factor

getU

public UpperTriangDenseMatrix getU()

Returns the upper triangular factor

getLU

public DenseMatrix getLU()

Returns the decomposition matrix

rcond

public double rcond(Matrix A,
                    Matrix.Norm norm)

Computes the reciprocal condition number, using either the infinity norm of the 1 norm.

Parameters:

A - The matrix this is a decomposition of

norm - Either Norm.One or Norm.Infinity

Returns:

The reciprocal condition number. Values close to unity indicate a well-conditioned system, while numbers close to zero do not.

getPivots

public int[] getPivots()

Returns the row pivots

isSingular

public boolean isSingular()

Checks for singularity

solve

public DenseMatrix solve(DenseMatrix B)
                  throws MatrixSingularException

Computes A\B, overwriting B

Throws:

MatrixSingularException

transSolve

public DenseMatrix transSolve(DenseMatrix B)
                       throws MatrixSingularException

Computes AT\B, overwriting B

Throws:

MatrixSingularException