Function: hgemm

Applicable Products

Product	Supported (√/x)
Ascend 950PR / Ascend 950DT	√
Atlas A3 training products / Atlas A3 inference products	√
Atlas A2 training products / Atlas A2 inference products	√
Atlas training products	√
Atlas inference products	√
Atlas 200I/500 A2 inference products	√

Function Usage

Computes matrix-matrix multiplication: C = αAB + βC. The inputs and outputs are all of data type aclFloat16. This API is asynchronous.

Prototype

C Prototype

        
         
           
           
             aclError aclblasHgemm(aclTransType transA,
aclTransType transB,
aclTransType transC,
int m,
int n,
int k,
const aclFloat16 *alpha,
const aclFloat16 *matrixA,
int lda,
const aclFloat16 *matrixB,
int ldb,
const aclFloat16 *beta,
aclFloat16 *matrixC,
int ldc,
aclComputeType type,
aclrtStream stream)

            

          

        
       

Python Function

        
             ret = acl.blas.hgemm(trans_a, trans_b, trans_c,
m, n, k,
alpha, matrix_a, lda,
matrix_b, ldb,
beta, matrix_c, ldc,
type, stream)

Parameter Description

Parameter	Description
trans_a	Int, whether matrix A is transposed.
trans_b	Int, whether matrix B is transposed.
trans_c	Int, whether matrix C is transposed. Currently, only aclTrans_N is supported.
m	Int, number of rows in matrix A and matrix C.
n	Int, number of columns in matrix B and matrix C.
k	Int, number of columns in matrix A and number of rows in matrix B.
alpha	Int, data pointer address of scalar alpha used for executing the multiplication operation.
matrix_a	Int, data pointer address of matrix A.
lda	Int, leading dimension of matrix A. For row-major transpose, lda is the number of columns of matrix A. This parameter is reserved and can only be set to –1.
matrix_b	Int, data pointer address of matrix B.
ldb	Int, leading dimension of matrix B. For row-major transpose, ldb is the number of columns of matrix B. This parameter is reserved and can only be set to –1.
beta	Int, data pointer address of scalar beta used for executing the multiplication operation.
matrix_c	Int, data pointer address of matrix C.
ldc	Int, leading dimension of matrix C. This parameter is reserved and can only be set to –1.
type	Int, computation precision.
stream	Int, stream where the operator is executed.

Return Value Description

Return Value	Description
ret	Int, error code: 0 on success; else, failure.

Reference

For the API call sequence, see API Call Sequence.

Parent topic: CBLAS Interfaces (blas)