Deep Neural Network Library (DNNL)  1.3.0
Performance library for Deep Learning
Public Member Functions | List of all members
dnnl::matmul::desc Struct Reference

Descriptor for a matmul primitive. More...

#include <dnnl.hpp>

Collaboration diagram for dnnl::matmul::desc:
Collaboration graph
[legend]

Public Member Functions

 desc (const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &dst_desc)
 Constructs a descriptor for a matmul primitive. More...
 
 desc (const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &bias_desc, const memory::desc &dst_desc)
 Constructs a descriptor for a matmul primitive. More...
 

Detailed Description

Descriptor for a matmul primitive.

Examples:
cpu_matmul_quantization.cpp, cpu_sgemm_and_matmul.cpp, and inference_int8_matmul.cpp.

Constructor & Destructor Documentation

◆ desc() [1/2]

dnnl::matmul::desc::desc ( const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc dst_desc 
)
inline

Constructs a descriptor for a matmul primitive.

Inputs:

Outputs:

Parameters
src_descMemory descriptor for source (matrix A).
weights_descMemory descriptor for weights (matrix B).
dst_descMemory descriptor for destination (matrix C).

◆ desc() [2/2]

dnnl::matmul::desc::desc ( const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc bias_desc,
const memory::desc dst_desc 
)
inline

Constructs a descriptor for a matmul primitive.

Inputs:

Outputs:

Parameters
src_descMemory descriptor for source (matrix A).
weights_descMemory descriptor for weights (matrix B).
dst_descMemory descriptor for destination (matrix C).
bias_descMemory descriptor for bias.

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