Roman Lutz

Plus it happens in 3.7 as I found out recently!

@MiroDudik it looks like @riedgar-ms wanted to continue with this back in April and #563 doesn't seem like it's going to be finished anytime soon, so shall we finish this...

> The idea behind the `validate_and_reformat` function is the following: > > The keys of `threshold_dict` represent the values of the sensitive feature column(s). If there is just one sensitive...

> There is a strange problem when testing `Thresholder` with a simple Pipeline (see code below/in one of the last commits). When testing in my local IDE (visual studio code...

> * Throughout the guide, I use a class I wrote for myself which produces plots that help gain insight into the data. (The class is called `PlotPredictionInfoClassifier`, and the...

All of what you explained makes sense to me @bramreinders97. I still prefer the explicit way of specifying the threshold just because it leaves no doubt as to what is...

This is perhaps naive for reasons I haven't quite thought through yet, but how about reading the targets and deciding if it's binary, multiclass, or regression based on that? ExponentiatedGradient...

I don't have the time to review this in detail right now, but I do want to point out that you may want to consider splitting parts of this PR...

> @romanlutz Yes I would love to make a user guide that builds on the example, that is probably better than a notebook. Never worked with `.rst` files and `literalinclude`...

@mnalk this is a pull request so you would have to either fork @bram49's fork of Fairlearn from which he created this PR, or you can alternatively wait for him...