TensorFlow 2.9.0-rc0 Released
The first release-candidate for TensorFlow's version 2.9.0 has been let out to the public.
Created on April 18|Last edited on April 19
Comment
TensorFlow 2.9.0-rc0 has been released for those of you wanting to play with the newest version of TensorFlow before it's ready for a stable release. This version was released last Tuesday on GitHub with a Twitter announcement putting the spotlight on it today.
Being a minor version upgrade, it's a fairly big update with some important new features, fixes, and other changes.
Changes in TensorFlow 2.9.0-rc0
For Keras functionality: Various ResNet models have been added, new optimizer functionality has been added (sending old optimizer code to legacy), some logging utilities have been changed and added, as well as other changes including bug fixes and an API removal.
For TensorFlow Lite: Support for additional data types and quantization, changes in TF Lite converter tools, as well as other changes.
Additional changes include: BoostedTrees going from deprecated in version 2.8 to completely removed in version 2.9 (users should switch to using TensorFlow Decision Forests), some changes within tf.function, and a variety of other fixes and changes.
Find out more
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.