A descriptor of a Batch Normalization operation. More...

#include <dnnl_types.h>

Collaboration diagram for dnnl_batch_normalization_desc_t:

Public Attributes
dnnl_primitive_kind_t	primitive_kind

dnnl_prop_kind_t	prop_kind

dnnl_memory_desc_t	data_desc
	Source and destination memory descriptor. More...

dnnl_memory_desc_t	diff_data_desc
	Source and destination gradient memory descriptor. More...

dnnl_memory_desc_t	data_scaleshift_desc

dnnl_memory_desc_t	diff_data_scaleshift_desc

dnnl_memory_desc_t	stat_desc

float	batch_norm_epsilon
	Batch normalization epsilon parameter. More...

unsigned	flags

Detailed Description

A descriptor of a Batch Normalization operation.

Member Data Documentation

float dnnl_batch_normalization_desc_t::batch_norm_epsilon

Batch normalization epsilon parameter.

dnnl_memory_desc_t dnnl_batch_normalization_desc_t::data_desc

Source and destination memory descriptor.

dnnl_memory_desc_t dnnl_batch_normalization_desc_t::data_scaleshift_desc

Scale and shift data and gradient memory descriptors.

Scaleshift memory descriptor uses 2D dnnl_nc format[2,Channels]. 1-st dimension contains gamma parameter, 2-nd dimension contains beta parameter.

dnnl_memory_desc_t dnnl_batch_normalization_desc_t::diff_data_desc

Source and destination gradient memory descriptor.

dnnl_memory_desc_t dnnl_batch_normalization_desc_t::diff_data_scaleshift_desc

unsigned dnnl_batch_normalization_desc_t::flags

dnnl_primitive_kind_t dnnl_batch_normalization_desc_t::primitive_kind

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

dnnl_prop_kind_t dnnl_batch_normalization_desc_t::prop_kind

dnnl_memory_desc_t dnnl_batch_normalization_desc_t::stat_desc

Statistics memory descriptor.

Statistics (mean or variance) descriptor use 1D dnnl_x format[Channels].

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