#include <dnnl_types.h>

Collaboration diagram for dnnl_matmul_desc_t:

Public Attributes
dnnl_primitive_kind_t	primitive_kind

dnnl_memory_desc_t	src_desc
	Source memory descriptor. More...

dnnl_memory_desc_t	weights_desc
	Weights memory descriptor. More...

dnnl_memory_desc_t	bias_desc
	Bias memory descriptor. More...

dnnl_memory_desc_t	dst_desc
	Destination memory descriptor. More...

dnnl_data_type_t	accum_data_type
	The accumulator data type. Initialized automatically. More...

Detailed Description

A descriptor of a matrix multiplication operation.

2D case: dst[m, n] = src[m, k] * weights[k, n] + bias[m, n]

3D case: dst[mb, m, n] = src[mb, m, k] * weights[mb, k, n] + bias[mb, m, n]

Member Data Documentation

dnnl_data_type_t dnnl_matmul_desc_t::accum_data_type

The accumulator data type. Initialized automatically.

dnnl_memory_desc_t dnnl_matmul_desc_t::bias_desc

Bias memory descriptor.

dnnl_memory_desc_t dnnl_matmul_desc_t::dst_desc

Destination memory descriptor.

dnnl_primitive_kind_t dnnl_matmul_desc_t::primitive_kind

The kind of primitive. Used for self-identifying the primitive descriptor. Must be dnnl_matmul.

dnnl_memory_desc_t dnnl_matmul_desc_t::src_desc

Source memory descriptor.

dnnl_memory_desc_t dnnl_matmul_desc_t::weights_desc

Weights memory descriptor.

The documentation for this struct was generated from the following file: