Matrix Multiplication

A primitive to perform matrix-matrix multiplication. More...

Classes
struct	dnnl::matmul
	Matrix multiplication (matmul) primitive. More...
struct	dnnl_matmul_desc_t
	A descriptor of a matrix multiplication operation. More...

Functions
dnnl_status_t DNNL_API	dnnl_matmul_desc_init (dnnl_matmul_desc_t *matmul_desc, const dnnl_memory_desc_t *src_desc, const dnnl_memory_desc_t *weights_desc, const dnnl_memory_desc_t *bias_desc, const dnnl_memory_desc_t *dst_desc)
	Initializes a matrix multiplication descriptor. More...

Detailed Description

A primitive to perform matrix-matrix multiplication.

The batched mode is supported with 3D tensors.

See also

Matrix Multiplication in developer guide

Function Documentation

dnnl_matmul_desc_init()

dnnl_status_t DNNL_API dnnl_matmul_desc_init	(	dnnl_matmul_desc_t *	matmul_desc,
		const dnnl_memory_desc_t *	src_desc,
		const dnnl_memory_desc_t *	weights_desc,
		const dnnl_memory_desc_t *	bias_desc,
		const dnnl_memory_desc_t *	dst_desc
	)

Initializes a matrix multiplication descriptor.

Inputs:

• src (dnnl_query_src_md, 0)
• weights (dnnl_query_weights_md, 0)
• bias (dnnl_query_weights_md, 1)

Outputs:

• dst (dnnl_query_dst_md, 0)

Parameters

matmul_desc	Output descriptor for matmul primitive.
src_desc	Source memory descriptor (matrix A)
weights_desc	Weights memory descriptor (matrix B)
bias_desc	Bias memory descriptor. Passing NULL, a zero memory descriptor, or a memory descriptor with format_kind set to dnnl_format_kind_undef disables the bias term.
dst_desc	Destination memory descriptor (matrix C).

Returns

dnnl_success on success and a status describing the error otherwise.