CosmoFast is a collection of differentiable cosmological modules, developed by He Jia and Uros Seljak. It is intended as an add-on package for BayesFast, but can also be used standalone. Feel free to contact He Jia if you would like to add your own modules to CosmoFast!
- BayesFast Website: https://www.bayesfast.org/
- Documentation: https://cosmofast.readthedocs.io/en/latest/
- Source Code: https://github.com/h3jia/cosmofast
- Bug Reports: https://github.com/h3jia/cosmofast/issues
We are upgrading BayesFast & CosmoFast to v0.2 with JAX, which would be faster, more accueate, and much easier to use than the previous version!
We plan to add pypi and conda-forge support later. For now, please install CosmoFast from source with:
git clone https://github.com/h3jia/cosmofast
cd cosmofast
pip install -e .
# you can drop the -e option if you don't want to use editable mode
# but note that pytest may not work correctly in this case
To check if CosmoFast is built correctly, you can do:
pytest # for this you will need to have pytest installed
CosmoFast requires python>=3.7, cython, extension-helpers, jax>=0.3, jaxlib>=0.3 and numpy>=1.17. Currently, it has been tested on Ubuntu and MacOS, with python 3.7-3.10.
- Planck 2018 likelihoods
cosmofast.planck_18: Plik Lite high-l TT & TTTEEE, Commander low-l TT, Simall low-l EE & BB, Smica lensing full & CMB marginalized. All of these likelihoods are rewritten using JAX. Some of them are diagonalized for better performance with BayesFast. - Dark Energy Survey Y1 3x2 likelihood
cosmofast.des_y1: coming soon. - Pantheon 2022 likelihood
cosmofast.pantheon_22: coming soon.
CosmoFast is distributed under the Apache License, Version 2.0.
If you find CosmoFast useful for your research, please consider citing our papers accordingly:
- He Jia and Uros Seljak, BayesFast: A Fast and Scalable Method for Cosmological Bayesian Inference, in prep (for posterior sampling)
- He Jia and Uros Seljak, Normalizing Constant Estimation with Gaussianized Bridge Sampling, AABI 2019 Proceedings, PMLR 118:1-14 (for evidence estimation)