DeepSpeed icon indicating copy to clipboard operation
DeepSpeed copied to clipboard
verified publisher icon microsoft

Check device count before running dist tests

Open HeyangQin opened this issue 2 years ago • 2 comments

HeyangQin avatar Feb 07 '23 20:02 HeyangQin

@tjruwase I reworked the previous PR. This PR would check GPU count against world size for all dist tests so it avoids issues like https://github.com/microsoft/DeepSpeed/issues/2733 and https://github.com/microsoft/DeepSpeed/issues/2482 for all the dist unit tests

HeyangQin avatar Feb 08 '23 18:02 HeyangQin

@mrwyattii Hi Michael, could you take a look at the updated PR? Thanks!

HeyangQin avatar Feb 09 '23 19:02 HeyangQin