`vbmc.K`/`vp.K` fixes - Githubissues

acerbilab / pyvbmc

PyVBMC: Variational Bayesian Monte Carlo algorithm for posterior and model inference in Python

https://acerbilab.github.io/pyvbmc/

BSD 3-Clause "New" or "Revised" License

114 stars 6 forks source link

`vbmc.K`/`vp.K` fixes #101

Closed Bobby-Huggins closed 2 years ago

Bobby-Huggins commented 2 years ago

Fixes issue #98 by getting rid of vbmc.K, since the number of variational components K is an attribute of vp.
Adds tests to ensure that optim_state["vp_K"] also tracks the number of components and matches vp.K.
Adds tests to ensure that vbmc.parameter_transformer, vbmc.vp.parameter_transformer, and vbmc.function_logger.parameter_transformer are all equal.

Bobby-Huggins commented 2 years ago

I've closed the PR temporarily, since I realized I would also like to check that vbmc.parameter_transformer is updated correctly: I believe it is, but it's another attribute that has some annoying redundancies (in vbmc, vp, and function_logger).