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

Descriptor for an elementwise backward propagation primitive. More...

#include <dnnl.hpp>

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

Public Member Functions

 desc (algorithm algorithm, const memory::desc &diff_data_desc, const memory::desc &data_desc, float alpha=0, float beta=0)
 Constructs a descriptor for an elementwise backward propagation primitive. More...
 

Detailed Description

Descriptor for an elementwise backward propagation primitive.

Examples:
cnn_training_f32.cpp, and cpu_cnn_training_bf16.cpp.

Constructor & Destructor Documentation

◆ desc()

dnnl::eltwise_backward::desc::desc ( algorithm  algorithm,
const memory::desc diff_data_desc,
const memory::desc data_desc,
float  alpha = 0,
float  beta = 0 
)
inline

Constructs a descriptor for an elementwise backward propagation primitive.

Inputs:

Outputs:

Parameters
algorithmElementwise algorithm kind.
diff_data_descDiff source and destination memory descriptors.
data_descSource memory descriptor.
alphaThe alpha parameter for the elementwise operation. Specific meaning depends on the algorithm.
betaThe beta parameter for the elementwise operation. Specific meaning depends on the algorithm.

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