Implementation of a Python MCMC gibbs-sampler with adaptive stepping
Project description
GAStimator
Implementation of a Python MCMC gibbs-sampler with adaptive stepping.
While this is a simple MCMC algorithm, it is robust and stable and well suited to high dimensional problems with many degrees of freedom and very sharp likelihood features. For instance kinematic modelling of datacubes with this code has been found to be orders of magnitude quicker than using more advanced affine-invariant MCMC methods.
Install
You can install GAStimator with pip install gastimator. Alternatively you can download the code here, navigate to the directory you unpack it too, and run python setup.py install.
It requires the following modules:
- numpy
- matplotlib
- plotbin
- joblib
Documentation
To get you started, see the walk through here: https://github.com/TimothyADavis/GAStimator/blob/master/documentation/GAStimator_Documentation.ipynb
Author & License
Copyright 2019 Timothy A. Davis
Built by Timothy A. Davis <https://github.com/TimothyADavis>. Licensed under
the GNU General Public License v3 (GPLv3) license (see LICENSE).
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for gastimator-0.5.1-py3-none-any.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 72f43d1557382c00b432e146f6f7d39be30cd6169de40b30cefaaf674de32530 |
|
| MD5 | 8e454ccd3ed3ee389e5f7568fc6fb2f9 |
|
| BLAKE2b-256 | 7feb47ab60c8677d1fdc9662a855e301a1b0a705bab01afe6d97055ffa32a495 |
