Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Fabric: do `set_epoch` for `batch_sampler.sampler` (#16841)
C
Carlos Mocholí committed
235e6922593fd150715254ea719419b78c0cba30
Parent: beced48
Committed by GitHub <[email protected]>
on 2/23/2023, 12:11:29 AM