You need to install the following python packages to run SteParSyn (e.g. using pip):
numpy scipy matplotlib astropy extinction emcee tqdm celerite george pygtc
The repository contains the following programs:
- SteParSyn.py: the main code
- convsyn.py: convolution code
- readstats.py: reads the Markov chains
- runone.sh: it calls SteParSyn.py
- runall.sh: it calls runone.sh
- readone.sh: it calls readstats.py
- readall.sh: it calls readone.sh
To install the code: clone the repo, install the dependencies, create missing dirs ... :
$ git clone https://github.com/hmtabernero/SteParSyn
$ pip3 install --user numpy scipy matplotlib astropy extinction emcee tqdm celerite george pygtc
$ cd SteParSyn
$ mkdir eSPECTRA BINOUT PLOTS GRIDS
Download the synthetic grid (Fe_MARCS.bin) from https://drive.google.com/file/d/1pkHVbj5rNvONJSqlMO4IvhW0UBN3Jvxf/view?usp=sharing
You should put it in the folder GRIDS
Give permissions to programs:
$ chmod +x *sh
Relevant directories:
- GRIDS: here you should put the grid of synthetic spectra
- MASKS: line masks files
- RANGES: your range files
- CONFIG: here you can control the priors and which parameters to fix, priors, etc
- OPT: option files (iterations, number of chains, ...)
- SPECTRA: input spectra
- eSPECTRA: best fit spectra (in ascii)
- BINOUT: pickle files containing the MCMC sampling
- PLOTS: corner plots
The repo contains an example to compute the stellar atmospheric parameters of a solar spectrum. To run the example:
$ ./runall.sh
To get the results:
$ ./readall.sh
If you want to analyse your own data you need to do as follows (I am assuming that the name of your star is "sun_vesta")
- Provide a txt file containing: wavelengths, fluxes, and its uncertainities (same as SPECTRA/sun_vesta.txt)
- Create a mask file with: line names, line centre, mask centre, and mask width (same format as MASKS/sun_vesta_masks.txt)
- Generate an option file with your parameter options: limits, priors, etc. (You can use CONFIG/sun_vesta.conf as is)
Add the following line to runall.sh:
./runone.sh sun_vesta hermes MCMC
Add also this line to readall.sh
./readone.sh sun_vesta
I will write a more complete manual in the near future.
Finally, please cite the SteParSyn paper if you use it in your reasearch, you can find it here: https://ui.adsabs.harvard.edu/abs/2022A%26A...657A..66T/abstract
If you have any doubts please do not hesitate to e-mail me.