WNUT17 Shared task eval command

26 views

Skip to first unread message

unread,

Jul 6, 2017, 2:59:13 PM7/6/17

to Workshop on Noisy User-generated Text (WNUT)

Hi,

If you're having trouble evaluating your data or checking the scores, the command we used to generate this was of the form:

cat ../submissions/drexel_flytxt | awk '{print $NF}'| paste emerging.test.annotated - | ../eval/wnuteval.py

You might need to amend the paths to wnuteval.py and to the submission input.

All the best,

Leon

Reply all

Reply to author

Forward

0 new messages