How to monitor Neural Engine usage on M1 macs?

I'm now running Tensorflow models on my Macbook Air 2020 M1, but I can't find a way to monitor the Neural Engine 16 cores usage to fine tune my ML tasks.

The Activity Monitor only reports CPU% and GPU% and I can't find any APIs available on Mach include files in the MacOSX 11.1 sdk or documentation available so I can slap something together from scratch in C.

Could anyone point me in some direction as to get a hold of the API for Neural Engine usage. Any indicator I could grab would be a start. It looks like this has been omitted from all sdk documentation and general userland, I've only found a ledger_tag_neural_footprint attribute, which looks memory related, and that's it.
Post not yet marked as solved Up vote post of rgolive Down vote post of rgolive
5.7k views
  • I have noticed that when training models using tensorflow-metal on my M1 MacBook Air, the GPU History window in Activity Monitor shows near 100% utilization, but I'd like to know if the Neural Engine is being utilized as well.

Add a Comment

Replies

I have yet to find a direct API to the neural engines in my limited exposure. It appears that the Metal interface to Tensorflow is indeed to the GPU only at this point.

Are there any updates available on this issue, please?

Access to something like this would be really great. A tab in activity monitor or even a floating graph would be perfect

A tab in activity monitor or even a floating graph would be perfect

If you have specific requirements like that, the best way to get them in front of the folks who work on the code is to file an enhancement request.

Please post your bug number, just for the record.

Share and Enjoy

Quinn “The Eskimo!” @ Developer Technical Support @ Apple
let myEmail = "eskimo" + "1" + "@" + "apple.com"

You can use this tool for monitoring your through terminal command - sudo asitop https://github.com/tlkh/asitop

  • Did you try to run it on Macos Ventura? I got errors and readme says that it only works on macOS Monterey. :(

Add a Comment