harmonic 1.0.0

Python package for efficient Bayesian evidence computation

Python package to efficiently compute the Learnt Harmonic Mean estimator of the Bayesian evidence

Harmonic is an open source and fully documented python implementation of the Learnt Harmonic Mean estimator for the Bayesian evidence or marginal likelihood. In practice one uses chains gathered separately through MCMC sampling software to train one of the Harmonic machine learning models which then stabilize the harmonic mean estimator.

Documentation

See comprehensive documentation at Harmonic Documentation.

Attribution

Please cite McEwen et al 2021 if this code package has been of use in any project. A link will be provided shortly upon submission. A BibTeX entry for the paper is:

@article{harmonic,
   author = {{McEwen}, J.~D. and {Wallis}, C.~G.~R. and {Price}, M.~A.},
    title = {Machine learning assisted marginal likelihood estimation:
            learnt harmonic mean estimator},
  journal = {Bayesian Analysis in prep},
     year = 2021
}

License

Harmonic is released under the GPL-3 license (see LICENSE.txt), subject to the non-commercial use condition (see LICENSE_EXT.txt)

harmonic
Copyright (C) 2021 Jason McEwen & contributors

This program is released under the GPL-3 license (see LICENSE.txt),
subject to a non-commercial use condition (see LICENSE_EXT.txt).

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.