search

mlcommons / cm4mlops

A collection of portable, reusable and cross-platform automation recipes (CM scripts) with a human-friendly interface and minimal dependencies to make it easier to build, run, benchmark and optimize AI, ML and other applications and systems across diverse and continuously changing models, data sets, software and hardware (cloud/edge)
http://docs.mlcommons.org/cm4mlops/
Apache License 2.0
7 stars 12 forks source link

Feature request: Apple/Metal support for Docker CM workflow #62

Open WarrenSchultz opened 4 weeks ago

WarrenSchultz commented 4 weeks ago

Title says it all :) It would be really helpful if there could be a workflow to use Docker on Macs for simplified testing. In my particular case, ResNet 50 v1.5, but any other models that are functional on Metal would be an added bonus.

arjunsuresh commented 3 weeks ago

Is it okay to do ResNet50 with tensorflow on Metal? If we need pytorch the tricky part is the current reference implementation doesnt support pytorch backend for ResNet50.

WarrenSchultz commented 3 weeks ago

Yeah, we're running it on TF already, so that wouldn't be a change for us, just a more easily repeatable environment.