koepsell / pyhmc Goto Github PK

==============================================
pyhmc: Hybrid Monte Carlo Sampling with python
==============================================

This package is a straight-forward port of the functions hmc2.m and
hmc2_opt.m from the MCMCstuff matlab toolbox written by Aki Vehtari
<http://www.lce.hut.fi/research/mm/mcmcstuff/>.
   
The code is originally based on the functions hmc.m from the netlab toolbox
written by Ian T Nabney <http://www.ncrg.aston.ac.uk/netlab/index.php>.

The portion of algorithm involving "windows" is derived from the C code for
this function included in the Software for Flexible Bayesian Modeling
written by Radford Neal <http://www.cs.toronto.edu/~radford/fbm.software.html>.
   
This software is distributed under the BSD License (see LICENSE file).

Authors
-------
- Kilian Koepsell <[email protected]>

Contents
--------
* README:
  This file.

* LICENSE:
  The files in this directory are distributed under the BSD license.

* hmc2.py:
  The main module with Hybrid Monte Carlo sampler.
  This is the only file needed, which can serve as a drop-in replacement
  of hmc2.m. Run some tests: 'python hm2.py'

* hmc2x.pyx:
  If cython is installed, this file will be used to speed up the main loop.

* f_energy.py:
  Some example energy functions with gradients.
  
* f_energyx.pyx:
  Cython version of some example energy functions with gradients.

* test_hmc2.py:
  A couple of tests. (requires nose)
  
* hmc.m, hmc2.m, hmc2_opt.m:
  Original matlab version as a reference.