Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Fix `torch.compile` patching when applied as decorator (#19627)
A
awaelchli committed
14e98ecbf21fdbc04369ddbadf72e4a51d876b1f
Parent: 7555384
Committed by GitHub <noreply@github.com>
on 3/15/2024, 12:12:48 PM