TensorFlow Logo is a Trademark of Google LLP.

HIGHLIGHTS:

  • Eager execution will be a central feature of 2.0.
  • Support for more platforms and languages, and improved compatibility.
  • Removed deprecated APIs and reduce the amount of duplication.

 

Released as an Open Source framework during the Google I/O 2015, TensorFlow is a machine learning framework/library created by Google that allows developers to leverage the emerging technology for their own uses. TensorFlow v1.0 only landed last year, the team behind the project has continued its work, just recently announcing that a preview for v2.0 of the platform would be coming later this year. This update could break compatibility with v1.0 APIs, but with its “focus on ease of use,” developers shouldn’t run into too much trouble updating.

Since the open-source release, TensorFlow has become the world’s most widely adopted machine learning framework, catering to a broad spectrum of users and use-cases. In this time, TensorFlow has evolved along with rapid developments in computing hardware, machine learning research, and commercial deployment.

Some highlights of what users can expect with TensorFlow 2.0:

  • Eager execution will be a central feature of 2.0. It aligns users’ expectations about the programming model better with TensorFlow practice and should make TensorFlow easier to learn and apply.

  • Support for more platforms and languages, and improved compatibility and parity between these components via standardization on exchange formats and alignment of APIs.

  • Removed deprecated APIs and reduce the amount of duplication, which has caused confusion for users.

There will also be a conversion tool for updating older v1.0 code to v2.0’s APIs. In cases where there is no API equivalent, a compatibility library for the older version will also be an option, though it won’t be a beneficiary of any future feature development after v2.0 is finalized, and security updates for the library are only promised for a year.

If you’re currently a developer using TensorFlow, you’ll want to make sure you update your apps or services to make use of v2.0 when it arrives. To that end, it may also be good to participate in the development process via the public comments, and follow along with things on the project’s mailing list.

Source: Google Groups

Leave a Reply