oneAPI Deep Neural Network Library (oneDNN)
Performance library for Deep Learning
1.96.0
dnnl_post_ops Struct Reference
oneDNN API » Primitives » Attributes

An opaque structure for a chain of post operations. More...

#include <dnnl_types.h>

Detailed Description

An opaque structure for a chain of post operations.

dnnl_post_ops can be used to perform some (trivial) operations like accumulation or eltwise after certain primitives like convolution.

Post operations might be combined together, making a chain of post operations. For instance one can configure convolution followed by accumulation followed by eltwise. This might be especially beneficial for residual learning blocks.

Warning
Of course not all combinations are supported, so the user should handle errors accordingly.

Supported post operations:

  • accumulation (base primitive: convolution)
  • eltwise (base primitive: convolution)
The documentation for this struct was generated from the following file: