Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Handle `set_to_none` when using DeepSpeed optimizer in Lite (#16275)
A
Adrian Wälchli committed
c65630712700e9075817337d9d416ebded4bfd2c
Parent: b195b7c
Committed by GitHub <[email protected]>
on 1/9/2023, 2:01:11 PM