Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Call TrainingTypePlugin collective functions directly instead of going through the Accelerator (#9677)
Co-authored-by: Adrian Wälchli <[email protected]> Co-authored-by: ananthsub <[email protected]> Co-authored-by: Rohit Gupta <[email protected]>
F
four4fish committed
15cd6ad45b9081c92be50042540eb8831b284cbd
Parent: ab06987
Committed by GitHub <[email protected]>
on 9/27/2021, 12:52:57 PM