Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Size:
1M - 10M
ArXiv:
License:
Download br/test/0000.parquet from unimelb-nlp/wikiann: direct link, hf CLI and curl.
- Browser
- Download file 65.4 kB
-
https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/95fd10b72732199a9910664c51c2627e974e7d15/br/test/0000.parquet
- Command line
-
hf download hf://datasets/unimelb-nlp/wikiann@95fd10b72732199a9910664c51c2627e974e7d15/br/test/0000.parquet
-
curl -L -o 0000.parquet https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/95fd10b72732199a9910664c51c2627e974e7d15/br/test/0000.parquet
65.4 kB
- Xet hash:
- f6792b31b38334d25bcc70f3c276189610fd9643946f3c3eb8677684ebcf17a3
- Size of remote file:
- 65.4 kB
- SHA256:
- 537b3239943fac1c735386adeadc526427c6eb1442b05ed9a05274813773c615
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.