A Bayesian hierarchical inference model for X-ray spectra with emission and/or absorption lines that may be shifted by some Doppler shift. Infers the number of Doppler shifts present in the data along with line intensities and widths.
Uses DNest4 and RJObject for reversible-jump MCMC.
Installing and Running the Model --------------------
First, make sure that DNest4 is installed. Also, make sure you have a variable DNEST4_PATH set to the directory that the DNest4 repository lives in, and have the DNest4/code/ directory in your PYTHONPATH variable.
You should now be able to go into ShiftyLines/code/ and type make to compile the model.
Run the model with
>>./main -d path/to/data/file -t 1
Use -d to set the path to the name of the file to be modelled. -t sets the number of parallel threads to be used: use -t 1 for a single thread, or more for parallel threads. For best results, choose something like -t 8 or something similar.
All content © 2016 the authors. The code is distributed under the GNU General Public license, version 3.
Pull requests are welcome! If you are interested in the further development of this project, please get in touch via the issues!