Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)  1.0.4
Performance library for Deep Learning
Public Attributes | List of all members
mkldnn_blocking_desc_t Struct Reference

Generic description of blocked data layout for most memory formats. More...

#include <mkldnn_types.h>

Public Attributes

mkldnn_dims_t strides
 The strides between the outermost blocks. More...
 
int inner_nblks
 The number of innermost blocks, e.g. 3 in case of OIhw_4i16o4i_
 
mkldnn_dims_t inner_blks
 The size of the blocks, e.g. {4, 16, 4} in case of OIhw_4i16o4i
 
mkldnn_dims_t inner_idxs
 The logical indices of the blocks, e.g. More...
 

Detailed Description

Generic description of blocked data layout for most memory formats.

See also
Understanding Memory Formats

Member Data Documentation

◆ strides

mkldnn_dims_t mkldnn_blocking_desc_t::strides

The strides between the outermost blocks.

In case of plain (non-blocked) formats the strides between dimensions.

◆ inner_idxs

mkldnn_dims_t mkldnn_blocking_desc_t::inner_idxs

The logical indices of the blocks, e.g.

{1, 0, 1} in case of 4i16o4i, because i is the 1st dim and o is the 0st dim


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