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
Hello, I have a model that I am trying to sample with a BetaBinomial likelihood. I describe the model in detail with the code in this discourse post. The error I am receiving is as follows:
BetaBinomial: the condition α > zero(α) is not satisfied.
I have inverse logit constraints in my model, so it is not possible for the alpha value to reach zero (though can be arbitrarily small). From my investigations it looks like this is an issue related to this issue with Distributions.jl. The developers of Gen.jl have identified this issue as well outlined here and here. It looks like their solution was rather straightforward. I am not sure if it is worth exploring something similar for Turing.jl (if it doesn't exist already).
The text was updated successfully, but these errors were encountered:
Hello, I have a model that I am trying to sample with a BetaBinomial likelihood. I describe the model in detail with the code in this discourse post. The error I am receiving is as follows:
I have inverse logit constraints in my model, so it is not possible for the alpha value to reach zero (though can be arbitrarily small). From my investigations it looks like this is an issue related to this issue with Distributions.jl. The developers of Gen.jl have identified this issue as well outlined here and here. It looks like their solution was rather straightforward. I am not sure if it is worth exploring something similar for Turing.jl (if it doesn't exist already).
The text was updated successfully, but these errors were encountered: