ZeroInflatedNegativeBinomial prior predictive sampling very slow

Hi all,

I have built a model that uses the RV ZeroInflatedNegativeBinomial. However, when doing a simple prior predictive sampling the model takes a long time to build and also the sampling is very slow. When using the RV NegativeBinomial instead, everything runs as expected.

Does anybody know why this is the case and how to fix this?

Currently I have manually implemented the zero inflation part and this also doesn’t has the above problems.

We are using pymc==4.1.4

Thanks for any help!

Can you share an example code snipet that exhibits the slow sampling that you are referring to?