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_exec_arg_t Struct Reference

An auxiliary structure to specify primitive's inputs/outputs at execution. More...

#include <mkldnn_types.h>

Collaboration diagram for mkldnn_exec_arg_t:
Collaboration graph
[legend]

Public Attributes

int arg
 An argument index, e.g. MKLDNN_ARG_SRC.
 
mkldnn_memory_t memory
 Input/output memory.
 

Detailed Description

An auxiliary structure to specify primitive's inputs/outputs at execution.

Warning
With this API it's impossible to preserve constness of memory, so all memories are passed w/o const qualifier. However only memories with output semantics might be changed during the execution
Examples:
cpu_cnn_inference_f32.c, cpu_cnn_training_f32.c, and gpu_getting_started.c.

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