Gamma method in initialization of Corr about pyerrors HOT 4 CLOSED

I agree with you, I am also in favor of changing this behavior. I would even say that it can be a bit dangerous to automatically call the gamma_method within say a fit routine as this will use the standard arguments for the gamma_method and overwrite any custom values for S or so which the user might have set before running the fit. This behavior could be confusing.

from pyerrors.

I agree as well. I only did it this way to avoid having to test for this in every single function.
If the performance cost is noticeable, it has to change.
One has to be a little careful here. Functions might rely on accurate dvalues and might not raise an error if it is zero.

from pyerrors.

Good point, hopefully our tests will catch these cases. I can take care of that today.

from pyerrors.

Very good, thanks! (And I fully agree with your concerns regarding silent calls of the gamma method)

from pyerrors.