BlasL3GeMM
Computes a matrix-matrix product with general matrices.
C = alpha * op(A) * op(B) + beta * C
where C is m-by-n matrix, op(A) is m-by-k matrix, op(B) is k-by-n matrix.
Method BlasL3GeMM is applied to the matrix A. Matrix A has size m-by-k if parameter transa='N', or k-by-m otherwise.
BLAS function GEMM.
Computing for type matrix<double>
bool matrix::BlasL3GeMM(
|
Computing for type matrix<float>
bool matrixf::BlasL3GeMM(
|
Computing for type matrix<complex>
bool matrixc::BlasL3GeMM(
|
Computing for type matrix<complexf>
bool matrixcf::BlasL3GeMM(
|
Parameters
transa
[in] Value from the ENUM_BLAS_TRANS enumeration, which specifies the operation:
if transa= 'N', then op(A) = A;
if transa= 'T', then op(A) = A**T;
if transa= 'C', then op(A) = A**H.
transb
[in] Value from the ENUM_BLAS_TRANS enumeration, which specifies the operation:
if transa= 'N', then op(B) = B;
if transa= 'T', then op(B) = B**T;
if transa= 'C', then op(B) = B**H.
alpha
[in] Scalar multiplier alpha.
B
[in] Matrix B of size k-by-n if transb='N', or of size n-by-k otherwise.
beta
[in] Scalar multiplier beta.
C
[in,out] Result matrix C of size m-by-n. If beta is not zero, then matrix C should contain actual data before entry.If matrix size differs from m-by-n, then matrix C will be resized and zeroed.
Return Value
Return true if successful, otherwise false in case of an error.
ENUM_BLAS_TRANS
An enumeration defining options op(A) and op(B).
ID |
Description |
|---|---|
BLASTRANS_N |
'N': No transpose |
BLASTRANS_T |
'T': Transpose |
BLASTRANS_C |
'C': Conjugate transpose |