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

Noisy likelihoods, small fixes #80

Closed Bobby-Huggins closed 2 years ago

Bobby-Huggins commented 2 years ago

Implements some smaller fixes as discussed on Slack:

  1. Default options (e.g. for acquisition functions and active sample VP/GP update) are now updated automatically when specifytargetnoise=True, as in MATLAB. (Options which have been set explicitly by the user are not updated.)
  2. No more ugly switch-case for acquisition functions; Acquisition function can now be specified either directly as the Python class instance, or as a string which is the name of the class.
  3. Acquisition function properties are now all in acq_info dict, instead of separate class properties.
  4. The quantile for VIQR/IMIQR acquisition functions can now be specified as an argument (default is 0.75).
  5. Added tests, other small changes.