Lukas

I would suggest you orient yourself along the lines of the normfactor code https://scikit-hep.org/pyhf/_modules/pyhf/modifiers/normfactor.html#normfactor the output shape of a modifier is ``` (n_modifiers, n_global_samples, n_alphas, n_global_bin) ``` here's the meaning...

Hi, I will likely take a stab at it since I need this for my analysis. (unless someone is already working on it?)

hi @mswiatlo - let's prioritize this then. is this the same type of formula that @balunas needs? and if so what do f(k),g(k),h(k) look like? Importantly - is each bin...

I think if the general structure is a multiplicative modifier per-sample that only depends on the parameters `sample_yield = base_sample_yield * f(POIs, NPs)` is definitely workable. The summing over samples...

thanks @alexander-held - so I guess that's compatible wiith my previous reply? can you confirm?

(not sure why github is not auto-linking this thread here: https://github.com/scikit-hep/pyhf/discussions/1627)

hi all, we're preparing our 0.7.0 release and with this we are making some progress on this front check out the code here https://gist.github.com/lukasheinrich/a7f23b71cf048727ca9012f5f6ea940a @alexander-held has been testing this and...

together with #1188 it should be much more straight forward to built a combine-like model

awesome that's a great start.. taking on the simplest example and successively adding features was also pyhf's approach in general. tagging also @clelange

Hi @jackaraz Can you point me to the MA5 Computation? What statistical model is used there?