dnnl::lrn_forward::desc Struct Reference

Descriptor for an LRN forward propagation primitive. More...

#include <dnnl.hpp>

Collaboration diagram for dnnl::lrn_forward::desc:

Public Member Functions
	desc (prop_kind prop_kind, algorithm algorithm, const memory::desc &data_desc, memory::dim local_size, float alpha, float beta, float k=1.f)
	Constructs a descriptor for a LRN forward propagation primitive. More...

Detailed Description

Descriptor for an LRN forward propagation primitive.

Examples:

cnn_inference_f32.cpp, cnn_training_f32.cpp, and cpu_cnn_training_bf16.cpp.

Constructor & Destructor Documentation

desc()

dnnl::lrn_forward::desc::desc ( prop_kind  prop_kind, algorithm  algorithm, const memory::desc &  data_desc, memory::dim  local_size, float  alpha, float  beta, float  k = 1.f  )

inline

Constructs a descriptor for a LRN forward propagation primitive.

Inputs:

• src (dnnl::primitive_desc_base::src_desc (0))

Outputs:

• dst (dnnl::primitive_desc_base::dst_desc (0))
• workspace (dnnl::primitive_desc_base::workspace_desc (0)), if alg_kind = dnnl::algorithm::pooling_max and prop_kind = dnnl::prop_kind::forward_training; must be queried for using dnnl::primitive_desc_base::query_md() after a corresponding primitive descriptor is created

Parameters

prop_kind	Propagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
algorithm	LRN algorithm kind: either dnnl::algorithm::lrn_across_channels, or dnnl::algorithm::lrn_within_channel.
data_desc	Source and destination memory descriptors.
local_size	Regularization local size.
alpha	The alpha regularization parameter.
beta	The beta regularization parameter.
k	The k regularization parameter.

---

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

• include/dnnl.hpp