Alexander Held

Assuming Gaussianity, the significance is `(best-fit POI value) / (POI uncertainty)`, e.g. for `mu = 1 +/- 0.5` this would translate to 2 sigma.

Maybe I misunderstood the way `strategy` is supposed to be used, but it seems like `None` is actually not a supported setting: ```python import pyhf model = pyhf.simplemodels.uncorrelated_background( signal=[12.0, 11.0],...

related issue for `staterror`: #1018

Hi, just to add an example of what other libraries do: `uproot4` can interpret cuts / selection requirements as python (https://github.com/scikit-hep/uproot4/blob/master/uproot4/language/python.py). I imagine that in practical applications, a very limited...

> yield = B + f(k1, k2, k3)*S1 + g(k1, k2, k3)*S2 + ... As long as `f` and `g` are not too complicated (I assume everything that can be...

@lukasheinrich I think so, yes. The `ROOT` version may be slightly less general as you have `f(POIs, NPs)` and I am not sure that you can do a dependency on...

I was curious about the possibility of converting datacards into pyhf workspaces and wrote a small utility https://github.com/alexander-held/datacard-to-pyhf. I do not know much about CMS Combine and the datacard format,...

Could a standalone Docker image also be possible? Having no CVMFS dependence at all would be useful to allow running validations anywhere.

Having the conda version available is great! I view a ready-to-use Docker image as complementary to that (I guess with conda there is compilation involved?).

@nucleosynthesis Yes, for the comparison with `pyhf` I was running `Combine` on lxplus. I saw small differences for [these example datacards](https://github.com/cms-analysis/HiggsAnalysis-CombinedLimit/tree/102x/data/tutorials/counting) when comparing a result obtained with `Combine` to a...