A descriptor of a Local Response Normalization (LRN) operation. More...
#include <dnnl_types.h>
Collaboration diagram for dnnl_lrn_desc_t:
Public Attributes | |
| dnnl_primitive_kind_t | primitive_kind |
| The kind of primitive. More... | |
| dnnl_prop_kind_t | prop_kind |
| The kind of propagation. More... | |
| dnnl_alg_kind_t | alg_kind |
| LRN algorithm. More... | |
| dnnl_memory_desc_t | data_desc |
| Source and destination memory descriptor. | |
| dnnl_memory_desc_t | diff_data_desc |
| Source and destination gradient memory descriptor. | |
| dnnl_dim_t | local_size |
| The number of channels to sum over (for cross-channel LRN) or the side length of the square region to sum over (for within-channel LRN). More... | |
| float | lrn_alpha |
| LRN alpha parameter. | |
| float | lrn_beta |
| LRN beta parameter. | |
| float | lrn_k |
| LRN k parameter. | |
Detailed Description
A descriptor of a Local Response Normalization (LRN) operation.
- Examples:
- cnn_inference_f32.c, and cpu_cnn_training_f32.c.
Member Data Documentation
◆ primitive_kind
| dnnl_primitive_kind_t dnnl_lrn_desc_t::primitive_kind |
The kind of primitive.
Used for self-identifying the primitive descriptor. Must be dnnl_lrn.
◆ prop_kind
| dnnl_prop_kind_t dnnl_lrn_desc_t::prop_kind |
The kind of propagation.
Possible values: dnnl_forward_training, dnnl_forward_inference, dnnl_backward, and dnnl_backward_data.
◆ alg_kind
| dnnl_alg_kind_t dnnl_lrn_desc_t::alg_kind |
LRN algorithm.
Possible values: dnnl_lrn_within_channel and dnnl_lrn_across_channels.
◆ local_size
| dnnl_dim_t dnnl_lrn_desc_t::local_size |
The number of channels to sum over (for cross-channel LRN) or the side length of the square region to sum over (for within-channel LRN).
The documentation for this struct was generated from the following file:
- include/dnnl_types.h