dotnet / dotnet/machinelearning
If evaluation data has only one class, how should we compute AUC?
Open
area-Core
enhancement
Priority:2
- Dominant language
- C#
- Stars
- 9.4k
- Forks
- 2k
- Avg merge
- 2d 20h
- Merged PRs (30d)
- 11
Description
Currently if there is only data from one class, we throw an exception (see issue #4692 ).
It might make more sense to return NaN instead of that.
Contributor guide
Assessment
This issue has not been assessed yet.