Deep Neural Network Library (DNNL)  1.1.3
Performance library for Deep Learning
Public Member Functions | List of all members
dnnl::layer_normalization_forward::desc Struct Reference

Descriptor for layer normalization forward propagation. More...

#include <dnnl.hpp>

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

Public Member Functions

 desc (prop_kind aprop_kind, const memory::desc &src_desc, const memory::desc &stat_desc, float epsilon, normalization_flags flags)
 Initializes a layer normalization descriptor for forward propagation using prop_kind (possible values are dnnl::forward_training and dnnl::forward_inference), memory descriptor data_desc, normalization parameter epsilon, and flags set using bit flags of type dnnl_layer_normalization_desc_t. More...
 

Detailed Description

Descriptor for layer normalization forward propagation.

Constructor & Destructor Documentation

◆ desc()

dnnl::layer_normalization_forward::desc::desc ( prop_kind  aprop_kind,
const memory::desc src_desc,
const memory::desc stat_desc,
float  epsilon,
normalization_flags  flags 
)
inline

Initializes a layer normalization descriptor for forward propagation using prop_kind (possible values are dnnl::forward_training and dnnl::forward_inference), memory descriptor data_desc, normalization parameter epsilon, and flags set using bit flags of type dnnl_layer_normalization_desc_t.

Note
In-place operation is supported; that is, dst points to the same memory as src.

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