Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)  1.0.4
Performance library for Deep Learning
Public Member Functions | List of all members
mkldnn::lrn_backward::desc Struct Reference

Descriptor for local response normalization backward propagation. More...

#include <mkldnn.hpp>

Collaboration diagram for mkldnn::lrn_backward::desc:
Collaboration graph
[legend]

Public Member Functions

 desc (algorithm aalgorithm, const memory::desc &data_desc, const memory::desc &diff_data_desc, memory::dim local_size, float alpha, float beta, float k=1.f)
 Initializes a descriptor for backward propagation using aalgorithm, memory descriptors data_desc and diff_data_desc, and regularization parameters local_size, alpha, beta, and k. More...
 

Detailed Description

Descriptor for local response normalization backward propagation.

Examples:
cpu_cnn_training_bf16.cpp, and cpu_cnn_training_f32.cpp.

Constructor & Destructor Documentation

◆ desc()

mkldnn::lrn_backward::desc::desc ( algorithm  aalgorithm,
const memory::desc data_desc,
const memory::desc diff_data_desc,
memory::dim  local_size,
float  alpha,
float  beta,
float  k = 1.f 
)
inline

Initializes a descriptor for backward propagation using aalgorithm, memory descriptors data_desc and diff_data_desc, and regularization parameters local_size, alpha, beta, and k.


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