The TensorFlow team is dedicated to bringing machine learning to every developer’s toolbox and specifically we are working very closely with Apple to support applied ML developers – engineers who need to build and deploy reliable, stable, performant ML systems, at any scale, and for the macOS platform: both x86 and the new Apple Silicon.
While TensorFlow works today on macOS x86, and both natively and under Rosetta on macOS Apple Silicon devices, we know there is more work to do to make the end user experience on Apple Silicon and to meet our goal to develop TensorFlow as the best-in-class platform for applied ML.
To help end users through this transition, we will work to streamline support across both TF Forums, the various GitHub repos and also on the Apple Developer forums (developer.apple.com)
We have created the following list of tags on TF Forums that we encourage you to use:
- mac_os - for anything macOS related
- apple_silicon - for anything specific to the new Apple Silicon devices (ie., M1, M2, etc.)
- pluggable_device - together with either mac_os or apple_silicon to clarify issues with the Apple Metal plugin