You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Adding the iminuit.Minuit instance to the FitResults container would allow interacting with that directly to access more of the iminuit functionality. This could be useful if users want to run other steps after a fit and would work similarly to return_result_obj=True in pyhf.infer.mle.fit.
The text was updated successfully, but these errors were encountered:
Hi @alexander-held, is this deeper than updating the class for FitResults and adding the instance returned from the fit? Just not sure what you intend by specifying that it would work similarly to the return_result_obj=True -- do you mean it should be optional?
Great, thanks for helping with this! Thinking about it a bit more, I see no downside to just always adding it to FitResults so we don't need to make this optional. The difference with pyhf is that the fit function there returns a varying amount of objects depending on its keyword arguments but we always only return a single FitResults object so users who do not need that Minuit instance can just ignore it.
Adding the
iminuit.Minuit
instance to theFitResults
container would allow interacting with that directly to access more of theiminuit
functionality. This could be useful if users want to run other steps after a fit and would work similarly toreturn_result_obj=True
inpyhf.infer.mle.fit
.The text was updated successfully, but these errors were encountered: