Bug description Using a custom dataloader I get the following error regardless of whether or not I set use_distributed_sampler=False TypeError: Lightning can't inject a (distributed) sampler into your batch sampler, because it doesn't su...
New issue Closed Description lxPreciousDeposits
trainer = Trainer(use_distributed_sampler=False) to disable Lightning attempting to inject the distributed sampler. And likewise, in Fabric it would be done in setup: fabric.setup_dataloaders(..., use_distributed_sampler=False) awaelchli added data handling and removed needs triage labels Aug ...