There are two prototypes of gemm
available, please see below.
gemm( const Scalar >, const MatrixA& a, const MatrixB& b, const Scalar >, MatrixC& c );
gemm( const Scalar alpha, const MatrixA& a, const MatrixB& b, const Scalar beta, MatrixC& c );
gemm
(short for generic
matrix-matrix operation) provides a C++ interface to BLAS routines SGEMM,
DGEMM, CGEMM, and ZGEMM. gemm
performs one of the matrix-matrix operations
C := alpha*op( A )*op( B ) + beta*C,
where op( X ) is one of
op( X ) = X or op( X ) = X' or op( X ) = conjg( X' ),
alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix.
The selection of the BLAS routine is done during compile-time, and is
determined by the type of values contained in type MatrixA
.
The type of values is obtained through the value_type
meta-function typename value_type<MatrixA>::type
. Table X below illustrates to
which specific routine this dispatching will take place.
Table 1.82. Dispatching of gemm.
Value type of MatrixA |
BLAS routine |
CBLAS routine |
CUBLAS routine |
---|---|---|---|
|
SGEMM |
cblas_sgemm |
cublasSgemm |
|
DGEMM |
cblas_dgemm |
cublasDgemm |
|
CGEMM |
cblas_cgemm |
cublasCgemm |
|
ZGEMM |
cblas_zgemm |
cublasZgemm |
The original routines SGEMM, DGEMM, CGEMM, and ZGEMM have thirteen arguments,
whereas gemm
requires
five arguments.
Defined in header boost/numeric/bindings/blas/level3/gemm.hpp
.
Parameters
The definition of term 1
The definition of term 2
The definition of term 3.
Definitions may contain paragraphs.
#include <boost/numeric/bindings/blas/level3/gemm.hpp> using namespace boost::numeric::bindings; blas::gemm( x, y, z );
this will output
[5] 0 1 2 3 4 5