Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Size:
1M - 10M
ArXiv:
License:
Download cs/test-00000-of-00001.parquet from unimelb-nlp/wikiann: direct link, hf CLI and curl.
- Browser
- Download file 755 kB
-
https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/main/cs/test-00000-of-00001.parquet
- Command line
-
hf download hf://datasets/unimelb-nlp/wikiann/cs/test-00000-of-00001.parquet
-
curl -L -o test-00000-of-00001.parquet https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/main/cs/test-00000-of-00001.parquet
755 kB
- Xet hash:
- 076d9ee23550135b2d636e39d9d933c87cdc8d19d48d6f5041ce1a95952744a4
- Size of remote file:
- 755 kB
- SHA256:
- 2ff92c1a1e3de7718ba0d30324efe5099e93c358e0c8e56040f695694d5c966a
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.