/pmclib

Population Monte Carlo (PMC) library

Primary LanguageC

pmclib

Population Monte Carlo (PMC) library

Information

Authors

Karim Benabed

Olivier Cappé, Jean-François Cardoso, Gersende Fort, Martin Kilbinger, Simon Prunet, Christian P. Robert, Darren Wraith

Version

1.1

Installation

Download the library from the github repository:

git clone https://github.com/martinkilbinger/pmclib

A directory pmclib is created automatically. Change into that directory, and install the code using the python script waf. First, configure the software, with optional installation path PATH.

cd pmclib
./waf configure --prefix=PATH

Note that waf is a python2 script.

You might need to set paths to required libraries and other options. waf can install libraries for you. Note that some libraries are not essential for the use of pmclib (but optional e.g. lapack, lua).

Next, compile the code:

./waf build

On success, install the library.

[sudo] ./waf install

This will copy include files to PATH/include/pmclib and PATH/include/pmctools, and the PMC libraries to PATH/lib. You might need to prepend sudo to waf if you want to install the software to a system-wide location. Otherwise, specify a local path in the configuration step above.

References

pmclib has been used for cosmological inference in Wraith, Kilbinger, Benabed et al. (2009). Please cite at least this paper when using pmclib. Cosmological model comparison using the Bayesian evidence estimated with PMC is discussed in Kilbinger, Wraith, Robert et al. (2010).