mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
MOTIVATION To generalize Distributed test cases for non-CUDA devices CHANGES - test/distributed/checkpoint/test_fsspec.py - test/distributed/checkpoint/test_state_dict.py - test/distributed/test_multi_threaded_pg.py Replaced hard coded device names with torch.accelerator.current_accelerator - torch/testing/_internal/distributed/_shard/sharded_tensor/__init__.py support for hccl backend Pull Request resolved: https://github.com/pytorch/pytorch/pull/156796 Approved by: https://github.com/guangyey, https://github.com/ezyang |
||
|---|---|---|
| .. | ||
| _internal | ||
| __init__.py | ||
| _comparison.py | ||
| _creation.py | ||
| _utils.py | ||