emcee#
emcee is an MIT licensed pure-Python implementation of Goodman & Weare’s Affine Invariant Markov chain Monte Carlo (MCMC) Ensemble sampler and these pages will show you how to use it.
This documentation won’t teach you too much about MCMC but there are a lot of resources available for that (try this one). We also published a paper explaining the emcee algorithm and implementation in detail.
emcee has been used in quite a few projects in the astrophysical literature and it is being actively developed on GitHub.
Basic Usage#
If you wanted to draw samples from a 5 dimensional Gaussian, you would do something like:
import numpy as np
import emcee
def log_prob(x, ivar):
return -0.5 * np.sum(ivar * x ** 2)
ndim, nwalkers = 5, 100
ivar = 1. / np.random.rand(ndim)
p0 = np.random.randn(nwalkers, ndim)
sampler = emcee.EnsembleSampler(nwalkers, ndim, log_prob, args=[ivar])
sampler.run_mcmc(p0, 10000)
A more complete example is available in the Quickstart tutorial.
How to Use This Guide#
To start, you’re probably going to need to follow the Installation guide to get emcee installed on your computer. After you finish that, you can probably learn most of what you need from the tutorials listed below (you might want to start with Quickstart and go from there). If you need more details about specific functionality, the User Guide below should have what you need.
We welcome bug reports, patches, feature requests, and other comments via the GitHub issue tracker, but you should check out the contribution guidelines first. If you have a question about the use of emcee, please post it to the users list instead of the issue tracker.
License & Attribution#
Copyright 2010-2021 Dan Foreman-Mackey and contributors.
emcee is free software made available under the MIT License. For details
see the LICENSE
.
If you make use of emcee in your work, please cite our paper (arXiv, ADS, BibTeX).
Changelog#
3.1.0 (2021-06-25)#
3.0.2 (2019-11-15)#
Added tutorial for moves interface
Added information about contributions to documentation
Improved documentation for installation and testing
Fixed dtype issues and instability in linear dependence test
Final release for JOSS submission
3.0.1 (2019-10-28)#
Added support for long double dtypes
Prepared manuscript to submit to JOSS
Improved packaging and release infrastructure
Fixed bug in initial linear dependence test
3.0.0 (2019-09-30)#
Added progress bars using tqdm.
Added HDF5 backend using h5py.
Added new
Move
interface for more flexible specification of proposals.Improved autocorrelation time estimation algorithm.
Switched documentation to using Jupyter notebooks for tutorials.
More details can be found on the docs.
2.2.0 (2016-07-12)#
Improved autocorrelation time computation.
Numpy compatibility issues.
Fixed deprecated integer division behavior in PTSampler.
2.1.0 (2014-05-22)#
Removing dependence on
acor
extension.Added arguments to
PTSampler
function.Added automatic load-balancing for MPI runs.
Added custom load-balancing for MPI and multiprocessing.
New default multiprocessing pool that supports
^C
.
2.0.0 (2013-11-17)#
Re-licensed under the MIT license!
Clearer less verbose documentation.
Added checks for parameters becoming infinite or NaN.
Added checks for log-probability becoming NaN.
Improved parallelization and various other tweaks in
PTSampler
.
1.2.0 (2013-01-30)#
Added a parallel tempering sampler
PTSampler
.Added instructions and utilities for using
emcee
withMPI
.Added
flatlnprobability
property to theEnsembleSampler
object to be consistent with theflatchain
property.Updated document for publication in PASP.
Various bug fixes.
1.1.3 (2012-11-22)#
Made the packaging system more robust even when numpy is not installed.
1.1.2 (2012-08-06)#
Another bug fix related to metadata blobs: the shape of the final
blobs
object was incorrect and all of the entries would generally be identical because we needed to copy the list that was appended at each step. Thanks goes to Jacqueline Chen (MIT) for catching this problem.
1.1.1 (2012-07-30)#
Fixed bug related to metadata blobs. The sample function was yielding the
blobs
object even when it wasn’t expected.
1.1.0 (2012-07-28)#
Allow the
lnprobfn
to return arbitrary “blobs” of data as well as the log-probability.Python 3 compatible (thanks Alex Conley)!
Various speed ups and clean ups in the core code base.
New documentation with better examples and more discussion.
1.0.1 (2012-03-31)#
Fixed transpose bug in the usage of
acor
inEnsembleSampler
.
1.0.0 (2012-02-15)#
Initial release.