BISP-1
Bayesian Inference of Starlight Polarization in 1D
BISP-1 implements the Bayesian method for tomographic decomposition of the plane-of-sky orientation of the magnetic field with the use of stellar polarimetry and distance developed in Pelgrims et al., 2022, A&A, (submitted). The method decomposes the polarization signal along distance in terms of dust polarization screens: dust clouds of the interstellar medium of our Galaxy.
BISP-1 has been developed in the framework of the PASIPHAE project.
The code can be cloned from the repository
git clone git@github.com:vpelgrims/Bisp_1.git /where/to/clone
and you can install the python library not (yet) using
pip install BISP-1
but in going to the folder /where/to/clone and running the command:
python -m pip install -e .
You may need sudo right.
Use the -e option at install time if you plan to make changes to BISP-1 without having to reinstall with each change. Otherwize, drop it. Don't forget the '.' at the end of the command, that is part of it.
Running the above command will deploy BISP-1 to your device and you can use the code from anywhere, not just the /where/to/clone folder.
The code uses the dynesty Python nested sampler to analyze the polarization + distance data through a maximum-likelihood method. The polarization signal is decomposed in thin dust-polarizing layers placed along distance. Each layer is characterized by its distance (formally the parallax) as well as by its mean polarization (Stokes parameters q and u) and intrinstic scatter which accounts for turbulence.
The model and the likelihood are fully determined in our paper.
BISP-1 depends on several libraries reported in the pyproject.toml file. They will be installed at installation if requirements are not met. BISP-1 does not explicitly rely on scipy but it calls utilities in dynesty that require it, so scipy is added as a requirement.
A jupyter notebook is provided to describe the overal structure of the code, to ease its handling, and demonstrate the several features that are included such as results display and plotting utilities. Assuming a star sample with appropriate formating, a basic usage with default setting could read as:
import BISP_1 as bisp
mystars = bisp.Stars(starsample) # initialization of a Stars object
mypriors = bisp.Priors(mystars) # initialization of a Priors object
mylos = bisp.Bisp(mystars,mypriors) # initialization of the main object
# 1. run the Bayesian decomposition assuming 1 cloud along the sightline:
mylos.run_OneLayer()
# 2. run the Bayesian decomposition assuming 2 clouds along the sightline:
mylos.run_TwoLayers()
# 3. compare the performance of both tested model
mylos.printSummaryStat()
If BISP-1 is useful for you and your research please cite this PASIPHAE paper:
"Starlight-polarization-based tomography of the magnetized ISM: Pasiphae's line-of-sight inversion method"
Pelgrims, Panopoulou, Tassis, Pavlidou et al., 2022 Astronomy & Astrophysics, submitted.
Bisp-1 -- ReadMe Copyright (C) 2022 V.Pelgrims