gemm

Description

This function performs the matrix-matrix multiplication : C = α * op(A) * op(B) + β * C
Type : polymorphic.

 

Documentation illustration (illustration unavailable in the archive).

Input parameters

 

Documentation illustration (illustration unavailable in the archive). A : class, 2D tensor of dimension M x K.
Documentation illustration (illustration unavailable in the archive). B : class, 2D tensor of dimension K x N.
Documentation illustration (illustration unavailable in the archive). C : class, 2D tensor of dimension M x N.
Documentation illustration (illustration unavailable in the archive). Op(A) : enum, operation that is non- or transpose.

    • CUBLAS_OP_N : the non-transpose operation is selected
    • CUBLAS_OP_T : the transpose operation is selected
    • CUBLAS_OP_H : the conjugate transpose operation is selected

Documentation illustration (illustration unavailable in the archive). Op(B) : enum, operation that is non- or transpose.

    • CUBLAS_OP_N : the non-transpose operation is selected
    • CUBLAS_OP_T : the transpose operation is selected
    • CUBLAS_OP_H : the conjugate transpose operation is selected

Documentation illustration (illustration unavailable in the archive). α : float, scalar used for multiplication.
Documentation illustration (illustration unavailable in the archive). β : float, scalar used for multiplication, if beta==0 then C does not have to be a valid input.

Output parameters

 

Documentation illustration (illustration unavailable in the archive). C : class, 2D tensor of dimension M x N.

Examples

All these examples are snippets PNG, you can drop these Snippet onto the block diagram and get the depicted code added to your VI (Do not forget to install Accelerator library to run it).

The non-transposed operation is selected for A and B

Documentation illustration (illustration unavailable in the archive).

The transposed operation is selected for A and B

Documentation illustration (illustration unavailable in the archive).

The transposed operation is selected for B

Documentation illustration (illustration unavailable in the archive).

The transposed operation is selected for A

Documentation illustration (illustration unavailable in the archive).