Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Languages:
English
Size:
10K - 100K
License:
Commit History
-
Remove unintuitive 'O-...' labels & increment version to 1.2.0
Tom Aarsena9c1360
-
Update README.md
d0f5cd2 -
Update fabner.py
2b23b17
-
Update README.md to reflect changes in the data loading script
da8cf68 -
Update fabner.py with new tagging formats
6577d73
-
Upload dataloading script and README.md
36655e7
-
initial commit
4eefd8c