Skip to content

KeyError when loading AbstentionBench: Musique dataset archive missing 'raw_data\\musique/test.json' #15

Description

@arjun0502

Description

When attempting to load the AbstentionBench dataset from HuggingFace, the download process fails with a KeyError related to the Musique dataset archive structure.

Error Message

KeyError: "There is no item named 'raw_data\\musique/test.json' in the archive"

Full traceback:
File ~.cache\huggingface\modules\datasets_modules\datasets\facebook--AbstentionBench\22fa3ef06b69d03d7e0abf64303b960f6e423906eb18f96e604f9851d1b9135d\AbstentionBench.py:95, in AbstentionBench.download_and_prepare(self, dl_manager, _)
...
KeyError: "There is no item named 'raw_data\\musique/test.json' in the archive

Steps to Reproduce

  1. Authenticate with HuggingFace:
    from huggingface_hub import login
    login()

  2. import datasets
    abstention_bench_data = datasets.load_dataset('facebook/AbstentionBench', trust_remote_code=True)

  3. Error occurs during the download/preparation phase when processing the Musique dataset

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions