ChainConsumer

ChainConsumer is a python package written to consume the output chains of Monte-Carlo processes and fitting algorithms, such as the results of MCMC. ChainConsumer's main function is to produce plots of the likelihood surface inferred from the supplied chain. In addition to showing the two-dimensional marginalised likelihood surfaces, marginalised parameter distributions are given, and maximum-likelihood statistics are used to present parameter constraints. In addition to this, parameter constraints can be output in the form of a LaTeX table. Finally, ChainConsumer also provides the functionality to plot the chains as a series of walks in parameter values, which provides an easy visual check on chain mixing and chain convergence.