File Exchange

image thumbnail

ParaMonte

version 1.4.1 (2.5 MB) by CDSLAB
ParaMonte: Plain Powerful Parallel Monte Carlo and MCMC Library for MATLAB, Python, Fortran, C++, C.

263 Downloads

Updated 16 Nov 2020

View Version History

GitHub view license on GitHub

Download the latest prebuilt READY-TO-USE ParaMonte MATLAB library from the GitHub release page for Windows OS:

https://github.com/cdslaborg/paramonte/releases/latest/download/libparamonte_MATLAB.zip

or for macOS / Linux:

https://github.com/cdslaborg/paramonte/releases/latest/download/libparamonte_MATLAB.tar.gz

For an illustration of the many powerful features of the library as well as serial and parallel example simulations see:

https://www.cdslab.org/paramonte/notes/examples/matlab/mlx/sampling_multivariate_normal_distribution_via_paradram.html

For more examples, see:

https://www.cdslab.org/paramonte/notes/examples/matlab/mlx/

Interested in receiving updates? Star and watch the GitHub repository of the library on GitHub:

https://github.com/cdslaborg/paramonte

If you find this package useful for your work, please rate it here and cite the ParaMonte library as described here:

https://www.cdslab.org/paramonte/notes/overview/preface/#how-to-acknowledge-the-use-of-the-paramonte-library-in-your-work

ParaMonte is a serial/parallel library of Monte Carlo simulation routines for stochastic optimization, sampling, and integration of mathematical objective functions of arbitrary-dimensions, in particular, the posterior probability distributions of Bayesian regression models in data science, Machine Learning, and scientific inference, with the design goal of unifying the automation (of Monte Carlo simulations), user-friendliness (of the library), accessibility (from multiple programming environments), high-performance (at runtime), and scalability (across many parallel processors).

The ParaMonte library currently includes ParaDRAM: a comprehensive implementation of the Delayed-Rejection Adaptive Metropolis-Hastings Markov Chain Monte Carlo (DRAM) sampler for both serial and parallel simulations in the MATLAB environment. In particular, the ParaMonte library enables you to run your simulations in parallel **without writing a single-line of parallel code** in the MATLAB environment, without even requiring the MATLAB parallelization toolbox.

The ParaMonte library has been designed to be blazing-fast while maintaining a high level of flexibility and user-friendliness.

The ParaMonte library is currently readily accessible from Python, MATLAB, Fortran, C++/C programming languages. For more information on the installation, usage, and examples, visit:

https://www.cdslab.org/paramonte

A pure-MATLAB implementation of the ParaDRAM algorithm of ParaMonte is also available as a separate library (named MatDRAM) here:

https://www.mathworks.com/matlabcentral/fileexchange/80866-matdram-delayed-rejection-adaptive-metropolis-mcmc

MATLAB Release Compatibility:

This software has been only tested with MATLAB R2018a and above. However, it should be compatible with MATLAB >=R2016b. If you find incompatibilities with any of the MATLAB releases newer than R2016a, please let us know by opening an issue on the GitHub issues page:

https://github.com/cdslaborg/paramonte/issues

This software is ready to use on x64-architecture computers (almost all of the recently-built computers are x64). If your platform is other than x64 or other than Windows/Linux/macOS, follow the simple guidelines here:

https://www.cdslab.org/paramonte/notes/installation/matlab/

to build the library for your local machine. Please let us also know at:

https://github.com/cdslaborg/paramonte/issues

so that we can support your platform and architecture in the future.

Cite As

See this page: https://www.cdslab.org/paramonte/notes/overview/preface/#how-to-acknowledge-the-use-of-the-paramonte-library-in-your-work

Comments and Ratings (10)

Joshua Osborne

Ahoo Abdi

fantastic! This package provides the ultimate Monte Carlo simulation environment and MCMC sampler that I have been looking for, from the automated simulation setup to the postprocessing and visualization of the results. It has the simulation restart ability which is great in case the simulation gets interrupted at any point before the completion. What has been the most useful in my research is the possibility of running the code in parallel to speed up the simulation and the fact that the package took care of all the parallelization without needing me to do anything in parallel.

Fatima B

Fatima B

M Hossenfelder

M Hossenfelder

Omid Bagheri

Brian Holmes

A King

A King

MATLAB Release Compatibility
Created with R2020a
Compatible with R2016b and later releases
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

example

example/himmelblau/MATLAB

example/mvn/MATLAB

src/interface/MATLAB/paramonte/auxil/classes

src/interface/MATLAB/paramonte/auxil/functions

src/interface/MATLAB/paramonte/interface

src/interface/MATLAB/paramonte/interface/@ParaDRAM

src/interface/MATLAB/paramonte/interface/@ParaMonteSampler

src/interface/MATLAB/paramonte/interface/@paramonte

src/interface/MATLAB/paramonte/kernel

src/interface/MATLAB/paramonte/kernel/@ParaDRAM_class

src/interface/MATLAB/paramonte/stats

src/interface/MATLAB/paramonte/vis

src/interface/MATLAB/paramonte/vis/cold

src/interface/MATLAB/paramonte/vis/colornames

src/interface/MATLAB/paramonte/vis/export_fig

src/interface/MATLAB/test