We found an issue in the prior draw additions in setup_sampler .
If we include constant red noise models, then line 1168 would think that there's red noise to be drawn from and include a list of priors to draw from. This results in an error because the list is empty.
I've added an extra check in each if statement to see if the list is not empty
We found an issue in the prior draw additions in
setup_sampler.If we include constant red noise models, then line 1168 would think that there's red noise to be drawn from and include a list of priors to draw from. This results in an error because the list is empty.
I've added an extra check in each if statement to see if the list is not empty
Found with @levischult and @stevertaylor