DongChunGuang
Results
1
comments of
DongChunGuang
@sgugger I have located the problem statement: ``` trainer.py def _pad_across_processes(self, tensor, pad_index=-100): .... # Gather all sizes size = torch.tensor(tensor.shape, device=tensor.device)[None] sizes = self._nested_gather(size).cpu() .... ``` Stuck at .cpu()...