Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
English
ArXiv:
DOI:
Libraries:
Datasets
Dask
License:

Split by languages?

#7
by mhenrichsen - opened

Any overview of how many tokens each language in the dataset has?

it is english only

HuggingFaceFW org

Indeed, while there may be some non english text interleaved with english, the dataset is primarily english

is there a way to download just a subset of this without the whole library ?
Also ie for selectivly searching or extracting a subset what search terms should be used ie for (medical data only?)

HuggingFaceFW org

We added sample versions of the dataset, see the dataset card for more info

guipenedo changed discussion status to closed

Sign up or log in to comment