Spaces:
Runtime error
My dataset can not be evaluated
I can't select my dataset https://huggingface.co/datasets/ejschwartz/oo-method-test for evaluation. If I click "Evaluate models" from the model page, it initially redirects to https://huggingface.co/spaces/autoevaluate/model-evaluator?dataset=ejschwartz%2Foo-method-test but finally lands on https://huggingface.co/spaces/autoevaluate/model-evaluator?dataset=acronym_identification. If I type eschwartz in the dataset field, there is no autocompletion, and it refuses to accept a non auto-completed entry.
I forked the space and did some debugging, and the problem appears to be that https://huggingface.co/spaces/autoevaluate/model-evaluator/blob/main/requirements.txt requires an ancient version of huggingface-hub that does not understand pagination, and so only returns the first 10000 datasets.
Thank you, @ejschwartz . Could you please submit a PR to us here? Did you only update the huggingface-hub version, or did you also make other changes?
Thank you, @ejschwartz . Could you please submit a PR to us here? Did you only update the huggingface-hub version, or did you also make other changes?
I only updated the huggingface-hub version. The space did not actually work, but I assume that's because I didn't have some of the right secrets...
I'll file a PR.
Done! #47
PR was merged and issue is fixed. Hurray!