search

scikit-hep / pyhf

pure-Python HistFactory implementation with tensors and autodiff
https://pyhf.readthedocs.io/
Apache License 2.0
276 stars 82 forks source link

Renaming test statistics to use the mu subscript more explicitly #1246

Open kratsg opened 3 years ago

kratsg commented 3 years ago

Maybe too late to for such a change, but what do you think about using these names "q0", "qmu" and "qmu_tilde" in the hypotest test_stat argument? I believe they are more readable and the connection to the formulas in the asymptotics paper is even more clear with the "mu" subscript.

_Originally posted by @alexander-held in https://github.com/scikit-hep/pyhf/pull/1232#discussion_r553843992_

lukasheinrich commented 3 years ago

qµ qµ~ q0 or are unicode chars too much?

alexander-held commented 3 years ago

The unicode also looks good to me. The main motivation is that "qmu" and "qµ" read the same way as the variable in the paper, but "q" does not.

matthewfeickert commented 3 years ago

The unicode also looks good to me. The main motivation is that "qmu" and "qµ" read the same way as the variable in the paper, but "q" does not.

I agree with @alexander-held here w.r.t. trying to keep similarity with the paper names, though I think that if we're wanting to break with the paper and introduce more general notation that can be done if we want to give a lot of thought to it. Maybe something to save for an API breaking change in the future(?).

qµ qµ~ q0 or are unicode chars too much?

I'm a bit curmudgeonly on using unicode a lot, but maybe I need to get over that. :P So while I don't think I would enthusiastically vote for it, I also wouldn't fight it.

lukasheinrich commented 3 years ago

i'm fine with not using unicode (or only as an alias).. i sharer the same concerns

matthewfeickert commented 3 years ago

i'm fine with not using unicode (or only as an alias)

Oh I do like the idea of as an alias though. :+1: