Skip to content

Gradient Informed, GPU Accelerated Lens modelling (GIGALens) -- a package for fast Bayesian inference on strong gravitational lenses.

License

Notifications You must be signed in to change notification settings

parlange/gigalens

 
 

Repository files navigation

GIGA-Lens

Latest PyPI version

Gradient Informed, GPU Accelerated Lens modelling (GIGA-Lens) is a package for fast Bayesian inference on strong gravitational lenses. For details, please see our paper. See here for our documentation.

Usage

Installation

GIGA-Lens can be installed via pip:

pip install gigalens

If pip notes an error after installation about conflicting dependencies, these can usually be safely ignored. If you wish to test the installation, tests can be run simply by running tox in the root directory.

If you don’t have access to institutional GPUs, one easy way is to use GPU on Google Colab. Please remember the very first cell should have !pip install gigalens. If you do have access to institutional GPUs, you can set up a notebook to run on GPU. For example, at NESRC, you can choose the kernel tensorflow-2.6.0, and include in the first cell: !pip install gigalens.

Requirements

The following packages are requirements for GIGA-Lens. However, !pip install gigalens is all you need to do. In fact, separately installing other packages can cause issues with subpackage dependencies. Some users may find it necessary to install PyYAML.

tensorflow>=2.6.0
tensorflow-probability>=0.15.0
lenstronomy==1.9.3
scikit-image==0.18.2
tqdm==4.62.0

The following dependencies are required by lenstronomy:

cosmohammer==0.6.1
schwimmbad==0.3.2
dynesty==1.1
corner==2.2.1
mpmath==1.2.1

Authors

GIGALens was written by Andi Gu.

About

Gradient Informed, GPU Accelerated Lens modelling (GIGALens) -- a package for fast Bayesian inference on strong gravitational lenses.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 89.0%
  • TeX 8.6%
  • Python 2.4%