skip to navigation
skip to content

multichain_mcmc 0.3

Multichain MCMC framework and algorithms

A simple framework based on PyMC for multichain MCMC algorithms.

Contains working implementations of:
  • DREAM/DREAM_ZS sampler : multichain_mcmc.dream.DreamSampler
  • Adaptive Metropolis Adjusted Langevin Algorithm (AMALA) sampler : multichain_mcmc.amala.AmalaSampler
See the sampler classes for details. AMALA sampler requires PyMC branch with gradient information support to function.
http://github.com/pymc-devs/pymc/tree/gradientBranch
 
File Type Py Version Uploaded on Size
multichain_mcmc-0.3.tar.gz (md5) Source 2010-06-21 3MB