Datasets:
Languages:
English
Multilinguality:
monolingual
Size Categories:
1M<n<10M
Language Creators:
expert-generated
Annotations Creators:
no-annotation
Source Datasets:
original
ArXiv:
License:
NonMatchingSplitsSizesError ?
#1
by
Data-drone
- opened
It looks like the kaggle dataset has been edited and made bigger again so now the num samples and shard lengths are wrong. What is the best fix?
I think this part should be edited:
https://huggingface.co/datasets/arxiv_dataset/blob/main/README.md?code=true#L54
I updated the dataset size in the README.md https://huggingface.co/datasets/arxiv_dataset/discussions/2 and merged on main
It's fixed now :)
severo
changed discussion status to
closed