Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)  1.0.4
Performance library for Deep Learning
Functions
Service functions

Functions

mkldnn_status_t MKLDNN_API mkldnn_set_verbose (int level)
 Sets verbosity level (print information to stdout). More...
 
mkldnn_status_t MKLDNN_API mkldnn_set_jit_dump (int enable)
 Enables or disables dumping of JIT-generated code. More...
 
const mkldnn_version_t MKLDNN_API * mkldnn_version ()
 Gets library version information. More...
 

Detailed Description

Function Documentation

◆ mkldnn_set_verbose()

mkldnn_status_t MKLDNN_API mkldnn_set_verbose ( int  level)

Sets verbosity level (print information to stdout).

Possible levels are:

  • 0 – no verbose output (default)
  • 1 – primitive information at execution
  • 2 – primitive information at creation and execution
Note
Dumping information might affect performance. This setting overrides the MKLDNN_VERBOSE environment variable.

◆ mkldnn_set_jit_dump()

mkldnn_status_t MKLDNN_API mkldnn_set_jit_dump ( int  enable)

Enables or disables dumping of JIT-generated code.

The enable parameter can be:

  • 0 – disable
  • any other value – enable
Note
This setting overrides the MKLDNN_JIT_DUMP environment variable.

◆ mkldnn_version()

const mkldnn_version_t MKLDNN_API* mkldnn_version ( )

Gets library version information.

Version information includes:

  • major – major version number
  • minor – minor version number
  • patch – patch release number
  • hash – git commit hash