[ad_1]
Over time, however, the need to take another evolutionary step has emerged and the Linux Foundation has been chosen for the ability to host large multi-stakeholder open source projects “with the right balance between organizational structure and the identification of specific solutions for projects“, says Chintala. Not least the fact that Azure, Nvidia and many other companies have made significant investments, including technical contributions and community building efforts. “PyTorch is an important part of their future. With PyTorch continuing to grow as a multi-stakeholder project, it’s time to move to a broader open source base“adds the developer.
A key research framework
The Facebook developers have been unofficially collaborating with the Ai community for the development of PyTorch since 2016 and the approach has always been open. The proof is that they are counted 2-400 collaborators worldwidewhich contributed to the development of 154 thousand projects. Understandably these numbers have made this framework one of the key solutions for Ai research. Moreover, it is used by 80% of researchers who present projects at machine learning conferences around the world. PyTorch has also become the basis of products made by Amazon Web Services, Microsoft Azure, OpenAi and many other companies and research institutes.
Now, however, with the birth of the PyTorch Foundation (and inclusion in the Linux Foundation), an open and transparent direction will be ensured for development. The community-based open source philosophy will remain unchangedbut thanks to the collaboration of large companies such as Amazon Web Services, Google Cloud and others, the project will be able to enjoy one solid governance, diversified leadership and additional investments.
“The foundation will strive to abide by four principles: staying open, maintaining its neutral nature, ensuring fairness, and forging a strong technical identity. One of its top priorities will be sand clearly learn corporate governance from the technical one of PyTorch“Meta points out in the official note
Obviously Meta will continue to invest in PyTorch and will be employed as core research framework e production in the Ai. The transition does not involve any changes in the developer code, key project, or operating models. The development contributors and licensing guidelines will also remain unchanged. All versions, features and technical direction will continue to be driven by the PyTorch community.
The PyTorch project especially in the last five years has gained visibility not only in academic but also industrial environments with commercial applications. Today there are several university courses at for example Stanford, New York University, École polytechnique fédérale de Lausanne in Switzerland and Oxford in the United Kingdom. While in industry the greatest impacts have materialized in the automotive sector with autonomous driving, in healthcare and in aerospace. Examples can be cited Tesla’s Autopilot and Uber’s Pyro machine learning framework.
.
[ad_2]
Source link
