Datasets:
Follow
Speech and Language Technology, DFKI 94
Sep 13, 2023 14:35 relaxed parsing of attributions This allows to parse attribution annotations where the target annotation is separated from the value by a TAB instead of a SPACE. Note, that this is not allowed by the BRAT-Standoff-Format (https://brat.nlplab.org/standoff.html). May 10, 2023 15:38 use config.data_dir instead of checking if url points to local directory this fixes "NotImplementedError: os.path.isdir is not extended to support URLs in streaming mode" when calling with a remote (zip) target as url. Oct 25, 2022 06:38 Fix task tags (#2 ) dfki-nlp +1: albertvillanova - Fix task tags (1ecc80fe9f2d33946574e5cc132f3380c9c5427d)
Co-authored-by: Albert Villanova <albertvillanova@users.noreply.huggingface.co> Jul 4, 2022 06:10 Fix `license` metadata (#1 ) dfki-nlp +1: julien-c - Fix `license` metadata (bbddf21087b7fec004488d524b0e1a39c2ce58ca)
Co-authored-by: Julien Chaumond <julien-c@users.noreply.huggingface.co> May 24, 2022 12:23 add example usage for brat script 10:07 Update README.md 10:07 remove languages 10:06 Create brat.py 10:06 Delete sciarg.py May 16, 2022 05:09 Update README.md 04:55 Update sciarg.py 04:54 Delete dataset_infos.json 04:54 Update README.md May 10, 2022 11:16 Update README.md 11:15 Update README.md 11:12 Upload README.md 10:53 Upload dataset_infos.json 10:53 Upload sciarg.py 06:13 initial commit