Package org.ejml.sparse.csc.mult

Class ImplMultiplication_MT_DSCC

java.lang.Object

org.ejml.sparse.csc.mult.ImplMultiplication_MT_DSCC

public class ImplMultiplication_MT_DSCC
extends Object

Implementation of concurrent matrix multiplication for DSCC matrices. Please invoke through CommonOps_MT_DSCC as that provides a nicer API.

Constructor Summary

Constructors

Constructor	Description
ImplMultiplication_MT_DSCC()

Method Summary

Modifier and Type	Method	Description
static void	mult(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, boolean add, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	mult(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	mult(DMatrixSparseCSC A, DMatrixSparseCSC B, DMatrixSparseCSC C, GrowArray<Workspace_MT_DSCC> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multAdd(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multAddTransA(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multAddTransAB(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multAddTransB(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multTransA(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multTransAB(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multTransB(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, boolean add, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	multTransB(DMatrixSparseCSC A, DMatrixRMaj B, DMatrixRMaj C, GrowArray<DGrowArray> listWork)	Invoke through CommonOps_MT_DSCC as it will manage the input contract
static void	stitchMatrix(DMatrixSparseCSC out, int numRows, int numCols, GrowArray<Workspace_MT_DSCC> listWork)	Combines results from independent blocks into a single matrix

Methods inherited from class java.lang.Object

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

Constructor Details

ImplMultiplication_MT_DSCC

public ImplMultiplication_MT_DSCC()

Method Details

mult

public static void mult(DMatrixSparseCSC A,
 DMatrixSparseCSC B,
 DMatrixSparseCSC C,
 GrowArray<Workspace_MT_DSCC> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

Performs matrix multiplication. C = A*B. The problem is broken up into as many "blocks" as there are threads available. Each block will process a set of columns independently. After running results from independent blocks are stitched together in the main thread. Extra storage requirements is about the same size as 'C'.

Parameters:

A - Matrix

B - Matrix

C - Storage for results. Array size is increased if needed.

listWork - Storage for internal workspace.

stitchMatrix

public static void stitchMatrix(DMatrixSparseCSC out,
 int numRows,
 int numCols,
 GrowArray<Workspace_MT_DSCC> listWork)

Combines results from independent blocks into a single matrix

mult

public static void mult(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multAdd

public static void multAdd(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

mult

public static void mult(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 boolean add,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multTransA

public static void multTransA(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multAddTransA

public static void multAddTransA(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multTransB

public static void multTransB(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multAddTransB

public static void multAddTransB(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multTransB

public static void multTransB(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C,
 boolean add,
 GrowArray<DGrowArray> listWork)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multTransAB

public static void multTransAB(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C)

Invoke through CommonOps_MT_DSCC as it will manage the input contract

multAddTransAB

public static void multAddTransAB(DMatrixSparseCSC A,
 DMatrixRMaj B,
 DMatrixRMaj C)

Invoke through CommonOps_MT_DSCC as it will manage the input contract