Model saved by torch2.1.0 can't load in 2.0.0

1

I save a torch model in torch 2.1.0 and try to load in 2.0.0, and I meet such error:

File "<stdin>", line 1, in <module>
  File "/home/app/.conda/envs/kuangxiu_dev/lib/python3.8/site-packages/torch/serialization.py", line 809, in load
    return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
  File "/home/app/.conda/envs/kuangxiu_dev/lib/python3.8/site-packages/torch/serialization.py", line 1172, in _load
    result = unpickler.load()
  File "/home/app/.conda/envs/kuangxiu_dev/lib/python3.8/site-packages/torch/_utils.py", line 172, in _rebuild_tensor_v2
    set_tensor_metadata(tensor, metadata)
  File "/home/app/.conda/envs/kuangxiu_dev/lib/python3.8/site-packages/torch/_utils.py", line 163, in set_tensor_metadata
    torch._C._set_tensor_metadata(tensor, metadata)  # type: ignore[attr-defined]
RuntimeError: Unexpected key `base_strides_/1536/1/` passed to setTensorMetadata.
2

These issues are expected since PyTorch is not forwards compatible but should be backwards compatible, i.e. loading an older checkpoint in newer PyTorch versions should work.