Package org.ejml.sparse.csc.mult

Class ImplMultiplication_FSCC

java.lang.Object

org.ejml.sparse.csc.mult.ImplMultiplication_FSCC

@Generated("org.ejml.sparse.csc.mult.ImplMultiplication_DSCC")
public class ImplMultiplication_FSCC
extends Object

Implementation of single thread sparse matrix multiplication. Most functions should be invoked through CommonOps_FSCC. This code is focused on the implementation and does not check the contract or properly initialize input data structures.

Constructor Summary

Constructors

Constructor	Description
ImplMultiplication_FSCC()

Method Summary

Modifier and Type	Method	Description
static void	addRowsInAInToC(FMatrixSparseCSC A, int colA, FMatrixSparseCSC C, int colC, int[] w)	Adds rows to C[*,colC] that are in A[*,colA] as long as they are marked in w.
static float	dotInnerColumns(FMatrixSparseCSC A, int colA, FMatrixSparseCSC B, int colB, @Nullable IGrowArray gw, @Nullable FGrowArray gx)	Computes the inner product of two column vectors taken from the input matrices.
static void	mult(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C)	Invoke through CommonOps_FSCC
static void	mult(FMatrixSparseCSC A, FMatrixSparseCSC B, FMatrixSparseCSC C, @Nullable IGrowArray gw, @Nullable FGrowArray gx)	Performs matrix multiplication.
static void	multAdd(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C)	Invoke through CommonOps_FSCC
static void	multAddColA(FMatrixSparseCSC A, int colA, float alpha, FMatrixSparseCSC C, int mark, float[] x, int[] w)	Performs the operation x = x + A(:,i)*alpha
static void	multAddTransA(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C, FGrowArray workArray)	Invoke through CommonOps_FSCC
static void	multAddTransAB(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C)	Invoke through CommonOps_FSCC
static void	multAddTransB(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C, FGrowArray workArray)	Invoke through CommonOps_FSCC
static void	multTransA(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C, FGrowArray workArray)	Invoke through CommonOps_FSCC
static void	multTransA(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C, FGrowArray workArray, FOperatorBinary op)	Invoke through CommonOps_FSCC
static void	multTransAB(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C)	Invoke through CommonOps_FSCC
static void	multTransB(FMatrixSparseCSC A, FMatrixRMaj B, FMatrixRMaj C, FGrowArray workArray)	Invoke through CommonOps_FSCC

Methods inherited from class java.lang.Object

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

Constructor Details

ImplMultiplication_FSCC

public ImplMultiplication_FSCC()

Method Details

mult

public static void mult(FMatrixSparseCSC A,
 FMatrixSparseCSC B,
 FMatrixSparseCSC C,
 @Nullable
 @Nullable IGrowArray gw,
 @Nullable
 @Nullable FGrowArray gx)

Performs matrix multiplication. C = A*B

Invoke through CommonOps_FSCC

Parameters:

A - Matrix

B - Matrix

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

gw - (Optional) Storage for internal workspace. Can be null.

gx - (Optional) Storage for internal workspace. Can be null.

multAddColA

public static void multAddColA(FMatrixSparseCSC A,
 int colA,
 float alpha,
 FMatrixSparseCSC C,
 int mark,
 float[] x,
 int[] w)

Performs the operation x = x + A(:,i)*alpha

NOTE: This is the same as cs_scatter() in csparse.

addRowsInAInToC

public static void addRowsInAInToC(FMatrixSparseCSC A,
 int colA,
 FMatrixSparseCSC C,
 int colC,
 int[] w)

Adds rows to C[*,colC] that are in A[*,colA] as long as they are marked in w. This is used to grow C and colC must be the last filled in column in C.

NOTE: This is the same as cs_scatter if x is null.

Parameters:

A - Matrix

colA - The column in A that is being examined

C - Matrix

colC - Column in C that rows in A are being added to.

w - An array used to indicate if a row in A should be added to C. if w[i] < colC AND i is a row in A[*,colA] then it will be added.

mult

public static void mult(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C)

Invoke through CommonOps_FSCC

multAdd

public static void multAdd(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C)

Invoke through CommonOps_FSCC

multTransA

public static void multTransA(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C,
 FGrowArray workArray)

Invoke through CommonOps_FSCC

multAddTransA

public static void multAddTransA(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C,
 FGrowArray workArray)

Invoke through CommonOps_FSCC

multTransA

public static void multTransA(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C,
 FGrowArray workArray,
 FOperatorBinary op)

Invoke through CommonOps_FSCC

multTransB

public static void multTransB(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C,
 FGrowArray workArray)

Invoke through CommonOps_FSCC

multAddTransB

public static void multAddTransB(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C,
 FGrowArray workArray)

Invoke through CommonOps_FSCC

multTransAB

public static void multTransAB(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C)

Invoke through CommonOps_FSCC

multAddTransAB

public static void multAddTransAB(FMatrixSparseCSC A,
 FMatrixRMaj B,
 FMatrixRMaj C)

Invoke through CommonOps_FSCC

dotInnerColumns

public static float dotInnerColumns(FMatrixSparseCSC A,
 int colA,
 FMatrixSparseCSC B,
 int colB,
 @Nullable
 @Nullable IGrowArray gw,
 @Nullable
 @Nullable FGrowArray gx)

Computes the inner product of two column vectors taken from the input matrices.

dot = A(:,colA)'*B(:,colB)

Invoke through CommonOps_FSCC

Parameters:

A - Matrix

colA - Column in A

B - Matrix

colB - Column in B

Returns:

Dot product