Ian

I don't feel like it is particularly more revealing than equalized odds, but it is a formulation I see discussed (potentially because of its inclusion in AIF360). The use-case is...

> I would be ok to include this. The only thing I don't like about it is that e.g. a very low FPR could 'make up' for a very high...

> @IanEisenberg are you still planning to get back to this? Please feel free to say so. If not, someone else may be interested in continuing the pull request. I'm...

I get the same error as above, and can reproduce it following the steps in the above comment. Similarly removed when using pandas 0.25.0 or higher.

Did you ever figure out the cause of this? Or a workaround? We are dealing with a similar issue.