pyLFI
is a Python toolbox for Bayesian parameter estimation in models with intractable likelihood functions. By using Likelihood-Free Inference (LFI) schemes, in particular Approximate Bayesian Computation (ABC), pyLFI
estimates the posterior distributions over model parameters.
pyLFI
presently includes the following methods:
- Rejection ABC
- MCMC ABC
- Post-sampling regression adjustment.
pyLFI
was created as a part of the author's Master thesis.
pyLFI
can be installed directly from PyPI:
$ pip install pylfi
Python
>= 3.8
Documentation can be found at pylfi.readthedocs.io.
Check out the Examples gallery in the documentation.
The repository uses continuous integration (CI) workflows to build and test the project directly with GitHub Actions. Tests are provided in the tests
folder. Run tests locally with pytest
:
$ python -m pytest tests -v