The S-Lang implementation of the affine-invariant MCMC Hammer
isis_emcee is an S-Lang implementation of the affine-invariant Markov chain Monte Carlo (MCMC) ensemble sampler proposed by Goodman & Weare (2010), implemented in Python (emcee) by Foreman-Mackey et al. (2013), which has then been implemented in the Interactive Spectral Interpretation System (ISIS) in S-Lang by M. A. Nowak, modified by T. Beuchert, L. Corrales, and M. Kuehnel. The S-Lang MCMC hammer (emcee) is included in the Remeis ISISscripts.
1- Install HEASoft: download the HEASOFT Software
gunzip -c heasoft6.19.tar.gz | tar xf -
cd heasoft-6.19/BUILD_DIR/
./configure
make
make install
2- Install ISIS:
wget http://space.mit.edu/cxc/isis/install-isis.sh
setenv HEADAS /usr/local/headas/x86_64-unknown-linux-gnu-libc2.7/
sh install-isis.sh DIR
3- Install ISISscripts: download the Remeis ISISscripts
or obtain the developing version from the ISISscripts git-repository:
git clone http://www.sternwarte.uni-erlangen.de/git.public/isisscripts
tar xfz isisscripts.tgz
Add it to to your personal ~/.isisrc startup file
add_to_isis_load_path("/path/to/isisscripts/");
Load the ISISscripts in your S-Lang script:
require("isisscripts");
4- Install the S-lang MCMC Hammer:
Obtain the developing version from the github:
git clone https://github.com/mcfit/isis_emcee.git
Load the S-lang MCMC Hammer in your S-Lang script:
require("isis_emcee");