Thank you, that’s great. I am not doubting that it is less efficient. Do you maybe have a reference where I could read up on that?
Maybe as a general remark, the popularity of Bayesian methods in many areas of physics is increasing rapidly. However, the evaluation of models in many of those areas takes non-negligible time and can easily become the bottleneck for the analysis. So, wherever possible, caching previous results to avoid the reevaluation would be great. I don’t know if this is at all possible, just wanted to leave this as a remark.
Thanks for the help!