Skip to content

Corrupted model checkpoint causing PytorchStreamReader failed reading zip archive error #2

Description

@mzc2113391

Hello,

I'm encountering a consistent error when trying to run inference on the following cases: HIS7, CAPSD, and B1LPA6. All three cases produce the same error, which suggests the model checkpoint might be corrupted.

Error traceback:
Traceback (most recent call last): File "main.py", line 444, in <module> sample(args) File "main.py", line 372, in sample ckpt = torch.load(os.path.join( File "/lustre/user/miniconda3/envs/proldm_env/lib/python3.8/site-packages/torch/serialization.py", line 1004, in load with _open_zipfile_reader(opened_file) as opened_zipfile: File "/lustre/user/miniconda3/envs/proldm_env/lib/python3.8/site-packages/torch/serialization.py", line 456, in __init__ super().__init__(torch._C.PyTorchFileReader(name_or_buffer)) RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
This error typically occurs when a PyTorch checkpoint file is corrupted or incomplete. It's likely that the shared model checkpoint has been corrupted during upload or storage.

Could you please verify the integrity of the current checkpoint and, if possible, upload a verified good copy to Google Drive? This would help resolve the issue for users trying to run inference on these cases.

Thanks for your help!

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions