Lukas

our REANA example has this requirement https://awesome-workshop.github.io/reproducible-analyses/07-higgstotautau-parallel/index.html cc @tiborsimko @clelange

adherence to NEP13 and NEP18 would make it useful to integrate jax into projects that rely on them for portability. Specifically we're looking to integrate jax w/ scale-out systems like...

as a minimal example this should work ``` pip install jax jaxlib numpy awkward` python >>> import awkward1 >>> import numpy as np >>> import jax.numpy as jnp >>> a...

A little video https://user-images.githubusercontent.com/2318083/201794664-0badc5f0-e2ea-4ec9-b3b3-5fa82dd852c6.mov

it's more customary to fit the NLL and I think we might have slight bus re errordef in minuit because we're fitting a scaled objective

not directly related but related in spirit: #956

reviving this based on some discussions with @pfackeldey @kratsg @matthewfeickert

I think, this is a duplicate of #1754

One subtle thing I noticed with newer versions of keras is that the argument order in the `binary_crossentropy` was reversed .. but fixing this I got the notebook working for...

we have `pyhf json2xml` which produces XML + TH1F stored in ROOT files so I guess it would be not too hard to interface to aghast