Neural Network Exchange Format Explained

Neural Network Exchange Format (NNEF)
Logo Size:	200px
Developer:	Khronos Group
Latest Release Version:	1.0.5
Latest Release Date:	^[1]
Operating System:	Cross-platform
Platform:	Cross-platform
Genre:	API

Neural Network Exchange Format (NNEF) is an artificial neural network data exchange format developed by the Khronos Group. It is intended to reduce machine learning deployment fragmentation by enabling a rich mix of neural network training tools and inference engines to be used by applications across a diverse range of devices and platforms.^{[2] [3]}

History

NNEF was proposed in 2015 by member companies of the Khronos Group as a device and implementation independent transfer format capable of describing any artificial neural net in terms of its structure, operations and data.

The first version of the standard was launched in provisional form in December 2017, and was ratified as an official Khronos standard in August 2018.

Objectives

The goal of NNEF is to enable data scientists and engineers to easily transfer trained networks from their chosen training framework into a wide variety of inference engines. NNEF encapsulates a complete description of the structure, operations and parameters of a trained neural network, independent of the training tools used to produce it and the inference engine used to execute it.

Governance and Availability

NNEF is maintained by the Khronos Group under its Open Governance Principles^[4] as follows:

• Any company is invited and able to join Khronos to contribute to and influence the development of its specifications;
• Finalized specifications are publicly and freely distributed at zero cost from the Khronos web-site;
• Any company can implement a Khronos specification and participating implementers can obtain a trademark license for conformant implementations and pay zero royalties to Khronos participants; and
• Developers may freely use implementations of Khronos specifications.

The NNEF specification is available on the Khronos NNEF registry and tools are available on Github

Versions

• NNEF 1.0 Provisional, Released 20 December 2017.^[5]
• NNEF 1.0, Released 13 August 2018^[6]
  • NNEF 1.0.1, Released 10 May 2019
  • NNEF 1.0.2, Released 13 July 2019^[7]

Industry Participation

The following Khronos members have participated in the NNEF working group:

Tools

The NNEF tools project on GitHub contains the following open source tools:

• File format Parser
• Bidirectional converters between NNEF and ONNX, Caffe, Caffe2, TensorFlow (python), TensorFlow (protobuf)
• Model zoo: reference collection of models converted to NNEF

See also

• Open Neural Network Exchange

Notes and References

1. Web site: Releases.
2. Web site: NNEF - Neural Network Exchange Format (NNEF). 2016-10-04. The Khronos Group. en. 2019-02-07.
3. Book: Seo. B.. Shin. M.. Mo. Y. J.. Kim. J.. 2018 International Conference on Information Networking (ICOIN) . Top-down parsing for Neural Network Exchange Format (NNEF) in TensorFlow-based deep learning computation . January 2018. 522–524. 10.1109/ICOIN.2018.8343173. 978-1-5386-2290-2. 5053900 .
4. https://www.khronos.org/members/ip-framework/ Khronos IP Framework
5. https://www.khronos.org/news/press/khronos-group-releases-nnef-1.0-standard-for-neural-network-exchange v1.0p Khronos PR
6. Web site: The Khronos Group launches new standard for deploying trained neural networks. 2018-08-13. SD Times. en-US. 2019-02-11.
7. Web site: Khronos NNEF Registry - The Khronos Group Inc. www.khronos.org. 2019-08-15.