Hi Team,
unfortunately, my submission fails with the following error:
```
WARNING: Your kernel does not support swap limit capabilities or the cgroup is not mounted. Memory limited without swap.
Traceback (most recent call last):
File "/tmp/codalab/tmpFR2Di3/run/program/score.py", line 187, in
x.eval(args.input_dir, args.output_dir)
File "/tmp/codalab/tmpFR2Di3/run/program/score.py", line 142, in eval
self.get_document_dict_by_system_id(subtask, JsonAnnotation, gold_ann).items()):
File "/tmp/codalab/tmpFR2Di3/run/program/score.py", line 56, in get_document_dict_by_system_id
gold_sent_offs = gold_annos[os.path.splitext(fn)[0]].sent_offs
KeyError: '19195_Tagesordnungspunkt_18_SPD_Thews_ID1919511500_26.11.2020'
```
A file with that name exists in the test set, as well as in my submission.zip.
Any Idea?
Posted by: Moiddes @ July 25, 2023, 2 p.m.Hi Moiddes,
yes, I think the reason for this is that your submission includes instances with empty cues but with predictions for roles (e.g., Message, Source etc.).
According to our schema, this shouldn't happen because each role has to be linked to a cue, meaning that you cannot have a role without a cue.
The error message seems a bit misleading though. But when I fix this problem in your files, then the eval script works fine.
Best,
Ines
Hi Ines,
thanks for clarification! Indeed the error message is a bit misleading but your explanation makes perfect sense.
I'll rework my predictions with that in mind!