Skip to main content

PyAutoLens

When two or more galaxies are aligned perfectly down our line-of-sight, the background galaxy is strongly lensed and appears multiple times or as an Einstein ring of light. PyAutoLens makes it simple to model strong gravitational lenses, like this one:

alt text

PyAutoLens is based on the following papers:

Adaptive Semi-linear Inversion of Strong Gravitational Lens Imaging

AutoLens: Automated Modeling of a Strong Lens's Light, Mass and Source

Python Example

With PyAutoLens, you can begin modeling a lens in just a couple of minutes. The example below demonstrates a simple analysis which fits a lens galaxy's light, mass and a source galaxy.

from autofit import conf
from autofit.optimize import non_linear as nl
from autolens.pipeline import phase as ph
from autolens.data.array import mask as msk
from autolens.model.galaxy import galaxy_model as gm
from autolens.data import ccd
from autolens.model.profiles import light_profiles as lp
from autolens.model.profiles import mass_profiles as mp
from autolens.data.plotters import ccd_plotters
from autolens.lens.plotters import lens_fit_plotters

import os

# In this example, we'll generate a phase which fits a simple lens + source plane system.

# Get the relative path to the data in our workspace and load the ccd imaging data.
path = '{}/../'.format(os.path.dirname(os.path.realpath(__file__)))

lens_name = 'example_lens'

ccd_data = ccd.load_ccd_data_from_fits(image_path=path + '/data/' + lens_name + '/image.fits', 
                                       psf_path=path+'/data/'+lens_name+'/psf.fits',
                                       noise_map_path=path+'/data/'+lens_name+'/noise_map.fits', 
                                       pixel_scale=0.1)

# Create a mask for the data, which we setup below as a 3.0" circle.
mask = msk.Mask.circular(shape=ccd_data.shape, pixel_scale=ccd_data.pixel_scale, radius_arcsec=3.0)

# We model our lens galaxy using a mass profile (a singular isothermal ellipsoid) and our source galaxy 
# a light profile (an elliptical Sersic). We load these profiles from the 'light_profiles (lp)' and 
# 'mass_profiles (mp)' modules.
lens_mass_profile = mp.EllipticalIsothermal
source_light_profile = lp.EllipticalSersic

# To setup our model galaxies, we use the GalaxyModel class, representing a galaxy the parameters of 
# which are variable and fitted for by the analysis.
lens_galaxy_model = gm.GalaxyModel(mass=lens_mass_profile)
source_galaxy_model = gm.GalaxyModel(light=source_light_profile)

# To perform the analysis, we set up a phase using the 'phase' module (imported as 'ph').
# A phase takes our galaxy models and fits their parameters using a non-linear search 
# (in this case, MultiNest).
phase = ph.LensSourcePlanePhase(lens_galaxies=dict(lens=gm.GalaxyModel(mass=mp.EllipticalIsothermal)),
                                source_galaxies=dict(source=gm.GalaxyModel(light=lp.EllipticalSersic)),
                                optimizer_class=nl.MultiNest, phase_name='example/phase_example')

# We run the phase on the ccd data, print the results and plot the fit.
result = phase.run(data=ccd_data)
lens_fit_plotters.plot_fit_subplot(fit=result.most_likely_fit)

Slack

We're building a PyAutoLens community on Slack, so you should contact us on our Slack channel before getting started. Here, I can introduce you to the community, give you the latest update on the software and discuss how best to use PyAutoLens for your science case.

Unfortunately, Slack is invitation-only, so first send me an email requesting an invite.

Features

PyAutoLens's advanced modeling features include:

  • Pipelines - build automated analysis pipelines to fit complex lens models to large samples of strong lenses.
  • Inversions - Reconstruct complex source galaxy morphologies on a variety of pixel-grids.
  • Adaption - (February 2019) - Adapt the lensing analysis to the features of the observed strong lens imaging.
  • Multi-Plane - (April 2019) Model multi-plane lenses, including systems with multiple lensed source galaxies.

HowToLens

Included with PyAutoLens is the HowToLens eBook, which provides an introduction to strong gravitational lens modeling with PyAutoLens. It can be found in the workspace and consists of 4 chapters:

  • Introduction - An introduction to strong gravitational lensing and PyAutolens.
  • Lens Modeling - How to model strong lenses, including a primer on Bayesian non-linear analysis.
  • Pipelines - How to build pipelines and tailor them to your own science case.
  • Inversions - How to perform pixelized reconstructions of the source-galaxy.

Workspace

PyAutoLens includes a workspace, which includes template scripts, examples and the HowToLens eBook. If you install PyAutoLens with conda or pip the workspace these should be included, however if not then workspace can be found on the master branch of this git repository.

If you install PyAutoLens with Docker a workspace will be generated for you in the home directory the first time you run the image. After the first time you run docker the workspace will persist any changes you make and won't be updated again.

Depedencies

PyAutoLens requires PyMultiNest and Numba.

Installation with conda

We recommend installation using a conda environment as this circumvents a number of compatibility issues when installing PyMultiNest.

First, install conda.

Create a conda environment:

conda create -n autolens python=3.7 anaconda

Install multinest:

conda install -c conda-forge multinest

Tell matplotlib what backend to use:

echo "backend : TKAgg" > ~/.matplotlib/matplotlibrc

Install autolens:

pip install autolens

Installation with pip

Installation is also available via pip, however there are reported issues with installing PyMultiNest that can make installation difficult, see the file INSTALL.notes

$ pip install autolens

Installation with Docker

An alternative to conda and pip is Docker, which makes installation easier by containerising the project.

If you don't have Docker then you can install it by following the guide here.

Once you have Docker installed you can download the PyAutoLens Docker project with the command:

docker pull autolens/autolens

This command can also be used to update the project.

The project can be run using:

docker run -it -e LOCAL_USER_ID=`id -u $USER` -h autolens -p 8888:8888 -p 6006:6006 -v $HOME/autolens_workspace:/home/user/workspace autolens/autolens

Once the project is running Docker will provide you with a URL. Copy and paste this URL into your browser, making sure you replace '(PyAutoLens or 127.0.0.1)' with '127.0.0.1'. This will bring up a Jupyter notebook including the 'howtolens' directory which is full of tutorials.

Any changes you make inside the Docker workspace will be saved in the autolens_workspace in your home directory.

Support & Discussion

If you're having difficulty with installation, lens modeling, or just want a chat, feel free to message us on our Slack channel.

Contributing

If you have any suggestions or would like to contribute please get in touch.

Publications

The following papers use PyAutoLens:

Galaxy structure with strong gravitational lensing: decomposing the internal mass distribution of massive elliptical galaxies

Novel Substructure and Superfluid Dark Matter

Credits

James Nightingale - Lead developer and PyAutoLens guru.

Richard Hayes - Lead developer and PyAutoFit guru.

Ashley Kelly - Developer of pyquad for fast deflections computations.

Nan Li - Docker integration & support.

Andrew Robertson - Critical curve and caustic calculations.

Andrea Enia - Voronoi source-plane plotting tools.

Metadata

Release files for autolens 0.9.16

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for autolens 0.9.16
File Size Uploaded Provenance Actions
autolens-0.9.16.tar.gz 366.9 kB Without provenance Details for autolens-0.9.16.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for autolens 0.9.16
File Interpreter ABI Platform Actions
autolens-0.9.16-py2.py3-none-any.whl Python 3, Python 2 none any Details for autolens-0.9.16-py2.py3-none-any.whl

Total release size: 933.3 kB

Release files / autolens-0.9.16.tar.gz

Download URL autolens-0.9.16.tar.gz
Size 366.9 kB
Tags Source
SHA-256 checksum
How to use checksums
0d8304c431f2360a6e74db11371ba2d06dd85f4a4ff1085210ffee2edc83dafb
BLAKE2b-256 checksum
How to use checksums
9268b3d55ac1bd12f97cf73142b68fe01ed55c33678b8a016ffd5ad82fd851d9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/33.1.1 requests-toolbelt/0.8.0 tqdm/4.26.0 CPython/2.7.15rc1

Release files / autolens-0.9.16-py2.py3-none-any.whl

Download URL autolens-0.9.16-py2.py3-none-any.whl
Size 566.4 kB
Tags Python 2 Python 3
SHA-256 checksum
How to use checksums
376feb5d85fd15150b3a3baca48c8dab7405b1b2527cc3792afee96d3e9fe185
BLAKE2b-256 checksum
How to use checksums
c75b4e5edc836daf11b78755c9ce3e0268a05df68c14a8415fc9c7d93bcd0ff4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/33.1.1 requests-toolbelt/0.8.0 tqdm/4.26.0 CPython/2.7.15rc1

Security for autolens 0.9.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.9.16 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page