Datasets:
Languages:
English
Multilinguality:
monolingual
Size Categories:
1M<n<10M
Language Creators:
found
Annotations Creators:
no-annotation
Source Datasets:
original
Tags:
License:
Dataset not streamable
#2
by
lhoestq
HF staff
- opened
Because the dataset script extracts a TAR archive and then globs inside it - this is not supported in streaming mode.
I'm working on fixing this by uploading the sharded data in the repository and using iter_archive in the dataset script/
done :)
lhoestq
changed discussion status to
closed