org.ejml.alg.block
Class BlockMultiplication

java.lang.Object
  org.ejml.alg.block.BlockMultiplication

public class BlockMultiplication
extends Object

Matrix multiplication for BlockMatrix64F. All sub-matrices must be block aligned.

Author:

Peter Abeles

Constructor Summary

BlockMultiplication()

Method Summary

static void	mult(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C) Performs a matrix multiplication on BlockMatrix64F submatrices. c = a * b
static void	multMinus(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C) Performs a matrix multiplication on BlockMatrix64F submatrices. c = c - a * b
static void	multMinusTransA(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C)
static void	multPlus(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C) Performs a matrix multiplication on BlockMatrix64F submatrices. c = c + a * b
static void	multPlusTransA(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C)
static void	multTransA(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C) Performs a matrix multiplication with a transpose on BlockMatrix64F submatrices. c = aT * b
static void	multTransB(int blockLength, D1Submatrix64F A, D1Submatrix64F B, D1Submatrix64F C) Performs a matrix multiplication with a transpose on BlockMatrix64F submatrices. c = a * b T

Methods inherited from class java.lang.Object

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

Constructor Detail

BlockMultiplication

public BlockMultiplication()

Method Detail

mult

public static void mult(int blockLength,
                        D1Submatrix64F A,
                        D1Submatrix64F B,
                        D1Submatrix64F C)

Performs a matrix multiplication on BlockMatrix64F submatrices.

c = a * b

It is assumed that all submatrices start at the beginning of a block and end at the end of a block.

Parameters:

blockLength - Size of the blocks in the submatrix.

A - A submatrix. Not modified.

B - A submatrix. Not modified.

C - Result of the operation. Modified,

multPlus

public static void multPlus(int blockLength,
                            D1Submatrix64F A,
                            D1Submatrix64F B,
                            D1Submatrix64F C)

Performs a matrix multiplication on BlockMatrix64F submatrices.

c = c + a * b

It is assumed that all submatrices start at the beginning of a block and end at the end of a block.

Parameters:

blockLength - Size of the blocks in the submatrix.

A - A submatrix. Not modified.

B - A submatrix. Not modified.

C - Result of the operation. Modified,

multMinus

public static void multMinus(int blockLength,
                             D1Submatrix64F A,
                             D1Submatrix64F B,
                             D1Submatrix64F C)

Performs a matrix multiplication on BlockMatrix64F submatrices.

c = c - a * b

It is assumed that all submatrices start at the beginning of a block and end at the end of a block.

Parameters:

blockLength - Size of the blocks in the submatrix.

A - A submatrix. Not modified.

B - A submatrix. Not modified.

C - Result of the operation. Modified,

multTransA

public static void multTransA(int blockLength,
                              D1Submatrix64F A,
                              D1Submatrix64F B,
                              D1Submatrix64F C)

Performs a matrix multiplication with a transpose on BlockMatrix64F submatrices.

c = a^T * b

It is assumed that all submatrices start at the beginning of a block and end at the end of a block.

Parameters:

blockLength - Size of the blocks in the submatrix.

A - A submatrix. Not modified.

B - A submatrix. Not modified.

C - Result of the operation. Modified,

multPlusTransA

public static void multPlusTransA(int blockLength,
                                  D1Submatrix64F A,
                                  D1Submatrix64F B,
                                  D1Submatrix64F C)

multMinusTransA

public static void multMinusTransA(int blockLength,
                                   D1Submatrix64F A,
                                   D1Submatrix64F B,
                                   D1Submatrix64F C)

multTransB

public static void multTransB(int blockLength,
                              D1Submatrix64F A,
                              D1Submatrix64F B,
                              D1Submatrix64F C)

Performs a matrix multiplication with a transpose on BlockMatrix64F submatrices.

c = a * b ^T

It is assumed that all submatrices start at the beginning of a block and end at the end of a block.

Parameters:

blockLength - Length of the blocks in the submatrix.

A - A submatrix. Not modified.

B - A submatrix. Not modified.

C - Result of the operation. Modified,