-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Experiment working, but some parsing errors in the document display in GUI #11
Comments
Hi, there seem to be two problems: The first one is that there is non-nested overlapping ground truth label included in your benchmark, namely this one: The other problem is that the benchmark labels (and label ids) are not sorted by their corresponding span start. I fixed these two problems and the benchmark was displayed just fine:
I realized that other benchmark readers don't enforce this sorting, which is something I need to fix. Thank you for reporting this in any case! |
The sorting of the ground truth labels for two other benchmark readers was implemented in 59f97bf. I'll close this issue. Feel free to reopen it in case your problem was not solved. |
Hello!
Happy to say I managed to run a full eval and view it in the GUI. Overall things are looking great.
There's just one thing, which might be an issue with my data but nevertheless might be worth looking into.
Most documents display properly, but a few have things like what is shown below (text and image).
Below that I have included the corresponding line from the
benchmark.jsonl
file.Any idea what is causing this? Thanks!
The text was updated successfully, but these errors were encountered: