Fix Mac mps dataloader_num_workers > 1 causes RuntimeError: _share_filename_: only available on CPU (#38819)

* Update trainer.py: add multiprocessing_context for mps devices

* Fix multiprocessing context for MPS with workers

* Apply style fixes

---------

Co-authored-by: Marc Sun <[email protected]>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

Author: 3 people

src/transformers/trainer.py

@@ -1023,12 +1023,16 @@ def _get_dataloader(
         else:
             data_collator = self._get_collator_with_removed_columns(self.data_collator, description=description)
 
+        # MPS requrires forking if multiple workers are specified
+        should_fork = torch.backends.mps.is_available() and self.args.dataloader_num_workers > 1
+
         dataloader_params = {
             "batch_size": batch_size,
             "collate_fn": data_collator,
             "num_workers": self.args.dataloader_num_workers,
             "pin_memory": self.args.dataloader_pin_memory,
             "persistent_workers": self.args.dataloader_persistent_workers,
+            "multiprocessing_context": "fork" if should_fork else None,
         }
 
         if not isinstance(dataset, torch.utils.data.IterableDataset):