Jammy2211 / autolens_subhalo

Results for Subhalo Scanning Paper

Scanning For Dark Matter Subhalos in Hubble Space Telescope Imaging of 54 Strong Lenses

This repository contains result and scripts to accompany the paper "Scanning For Dark Matter Subhalos in textit{Hubble} Space Telescope Imaging of 54 Strong Lenses":.

https://arxiv.org/abs/2209.10566

Structure

The workspace includes the following main directories:

• PyAuto: The PyAutoLens (and parent projects) source code used to generate the results and paper figures. Use

this via a GitHub clone to run the .sqlite files. - slacs: The results of fits to the SLACS sample and visualization scripts to produce paper plots. - bells: The results of fits to the BELLS-GALLERY sample and visualization scripts to produce paper plots. - subhalo: The scripts used to generate latex tables and figures in the paper.

What Results Are Here?

The slacs/output and bells/output folders contain folders with with following information for the lens light, mass and subhalo models fitted to every lens:

• model.info: The model parameterization and priors.
• model.results: The inferred model parameters (maximum likelihood, 1 and 3 sigma confidence intervals).
• search.summary: A summary of the _dynesty_ non-linear search.
• image: Images showing the results of the fit, triangle plots and other quantities.

What Results Are Not Here?

Due to GitHub file size limits the following results are not included:

• The Source pipeline results which perform model-fits initializing the source pixelization.
• The image folder of every fit of the subhale grid search (subhalo[2]_mass[total]_source_subhalo[search_lens_plane]).
• The ``dynesty` samples of every fit.

How Do I Get Those Results?

Sqlite3 databases containing all of the results, including the full Dynesty samples, can be found at the following Zenodo links:

https://zenodo.org/record/7051604#.YxjdFdTML-g

These databases are required to reproduce the majority of visuals in the paper via the scripts on this repository.