mxnet
Public Attributes | List of all members
mkldnn_pooling_desc_t Struct Reference

#include <mkldnn_types.h>

Collaboration diagram for mkldnn_pooling_desc_t:
Collaboration graph

Public Attributes

mkldnn_primitive_kind_t primitive_kind
 
mkldnn_prop_kind_t prop_kind
 
mkldnn_alg_kind_t alg_kind
 
mkldnn_memory_desc_t src_desc
 
mkldnn_memory_desc_t diff_src_desc
 
mkldnn_memory_desc_t dst_desc
 
mkldnn_memory_desc_t diff_dst_desc
 
mkldnn_dims_t strides
 
mkldnn_dims_t kernel
 
mkldnn_dims_t padding [2]
 
mkldnn_padding_kind_t padding_kind
 
mkldnn_data_type_t accum_data_type
 

Detailed Description

A descriptor of a pooling operation.

Member Data Documentation

mkldnn_data_type_t mkldnn_pooling_desc_t::accum_data_type

The accumulator data type. Initialized automatically.

mkldnn_alg_kind_t mkldnn_pooling_desc_t::alg_kind

The kind of pooling algorithm. Possible values: mkldnn_pooling_max and mkldnn_pooling_avg.

mkldnn_memory_desc_t mkldnn_pooling_desc_t::diff_dst_desc

Destination gradient memory descriptor.

mkldnn_memory_desc_t mkldnn_pooling_desc_t::diff_src_desc

Source gradient memory descriptor.

mkldnn_memory_desc_t mkldnn_pooling_desc_t::dst_desc

Destination memory descriptor.

mkldnn_dims_t mkldnn_pooling_desc_t::kernel

Pooling kernel spatial dimensions.

mkldnn_dims_t mkldnn_pooling_desc_t::padding[2]

Padding in each spatial dimension. padding[0] is a padding in the beginning (padding_l), padding[1] is a padding in the end (padding_r).

mkldnn_padding_kind_t mkldnn_pooling_desc_t::padding_kind

The kind of padding to use.

mkldnn_primitive_kind_t mkldnn_pooling_desc_t::primitive_kind

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

mkldnn_prop_kind_t mkldnn_pooling_desc_t::prop_kind

The kind of propagation. Possible values: mkldnn_forward_training, mkldnn_forward_inference, mkldnn_backward, and mkldnn_backward_data.

mkldnn_memory_desc_t mkldnn_pooling_desc_t::src_desc

Source memory descriptor.

mkldnn_dims_t mkldnn_pooling_desc_t::strides

Pooling kernel strides for spatial dimensions.


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