RR_fitter

Code to perform RR fitting method on spectra to recover fractional star formation histories and transmission functions

Getting Started

Download the entire RR_fitter folder to your local machine.

The input spectra to be reconstructed using the RR_fitter should be placed in input/name_of_galaxy_spectra_folder/

The input files (also contained in input/) should be edited to contain the name of the spectra files you wish to reconstruct, their paths and a series of optional parameters.

The input and setup files should not have their structure changed. The input and setup parameters are read in line by line, do not change the order of parameters in the files.

The setup file should be edited if you wish to change number of walkers used in the RR MCMC fit, the burn in period, the datacube of SSP models used to reconstruct the spectrum, etc.

The RR or fullshape fitters can then be run using

python run_RR_fitter.py input_file.txt setup_file.txt

Prerequisite python modules

• astropy
• corner
• matplotlib
• numpy
• scipy

Testing RR_fitter

To test the RR_fitter works, run on the example Lgal galaxies gal_spectrum_364_MGS_M10.fits and gal_spectrum_750_MGS_M10.fits using the provided example setup files. You can check you get the same output as that contained in the folder example_output.

Implement this test using the RR MCMC spectral fitter with the command

python run_RR_fitter.py input_Lgal_example.txt setup_Lgal_example.txt

and the below for the fullshape MCMC spectral fitter

python run_fullshape_fitter.py input_Lgal_example.txt setup_Lgal_example.txt

Authors

• Elizabeth Swann

License

This project is licensed under the MIT License - see the LICENSE.md file for details

Acknowledgments

• This code uses the emcee python package, created by Dan Foreman-Mackey and contributors, and is free software made avaliable under the MIT license (https://github.com/dfm/emcee). Note that this module is not listed as a prerequisite as it is contained as part of the RR_fitter in code/emcee and code/emcee_1 as slight changes have been made for use with RR_fitter.
• Thanks to Prof Peter Thomas and Dr Rita Tojeiro for all their contributions to this work.