search

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

Interface changes #114

Closed Bobby-Huggins closed 1 year ago

Bobby-Huggins commented 1 year ago
  1. Makes some minor changes to PyVBMC's interfaces: a. vbmc.optimize() now returns just two items: the vp and a results dictionary. The elbo, elbo_sd and success_flag are contained in the dictionary. b. There is now a vp.log_pdf() function for convenience (it calls vp.pdf() with log_flag=True and passes along all other arguments and keyword arguments. c. Adds tests for vp.log_pdf().
  2. Updates the documentation and examples with these changes.
  3. Adds separate .py files for each example, containing the full code. Links to these files are at the top of each example.
    • Note: The links won't work until this PR is merged because they point to the main branch (I couldn't get Sphinx to understand the relative links inside a Jupyter Notebook).
  4. Some other small changes to the documentation: a. Adds a "How does it work?" section to the README, with animated gif. b. Fixes BibTeX in GitHub pages to match README.
Bobby-Huggins commented 1 year ago

I haven't pushed the rebuilt docs yet, I'll wait until this PR is merged (though if anyone wants to view them sooner I can push them now).