ExtensibleMCMC.jl
[NOTE this package is currently under development]
- provides a modular implementation of the Markov chain Monte Carlo (MCMC) algorithm
- comes with essential functionality that you'd expect from your MCMC sampler
- is designed to be easily extensible by the users with their own specialized MCMC steps
A case in point is DiffusionMCMC.jl, in which ExtensibleMCMC.jl is used to define inference algorithms for diffusion processes.
Please see the documentation to learn more about this package.