Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Size:
1M - 10M
ArXiv:
License:
Download cs/train-00000-of-00001.parquet from unimelb-nlp/wikiann: direct link, hf CLI and curl.
- Browser
- Download file 1.52 MB
-
https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/main/cs/train-00000-of-00001.parquet
- Command line
-
hf download hf://datasets/unimelb-nlp/wikiann/cs/train-00000-of-00001.parquet
-
curl -L -o train-00000-of-00001.parquet https://huggingface.co/datasets/unimelb-nlp/wikiann/resolve/main/cs/train-00000-of-00001.parquet
1.52 MB
- Xet hash:
- eb1be6a7703775a25c88c58c0832836d0281d2f128ee6d2e2d2b3ee7a78ceed7
- Size of remote file:
- 1.52 MB
- SHA256:
- 06428c4a936622952ffcf39f11e7b9a834326523bd2dd8a043970eb7c38c9b8b
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.