SelfiSys is a Python package designed to address the issue of model misspecification in field-based, implicit likelihood cosmological inference.
It leverages the inferred initial matter power spectrum, enabling a thorough diagnosis of systematic effects in large-scale spectroscopic galaxy surveys.
- Custom hidden-box forward models
We provide a HiddenBox class to simulate realistic spectroscopic galaxy surveys. It accommodates fully non-linear gravitational evolution, and incorporates multiple systematic effects observed in real-world survey, e.g., misspecified galaxy bias, survey mask, selection functions, dust extinction, line interlopers, or inaccurate gravity solver.
- Diagnosis of systematic effects
Diagnose the impact of systematic effects using the inferred initial matter power spectrum, prior to performing cosmological inference.
- Cosmological inference
Perform inference of cosmological parameters using Approximate Bayesian Computation (ABC) with a Population Monte Carlo (PMC) sampler.
The documentation, including a detailed API reference, is available at hoellin.github.io/selfisys_public.
For practical examples demonstrating how to use SelfiSys, visit the SelfiSys Examples Repository.
- Tristan Hoellinger, tristan.hoellinger@iap.fr
Principal developer and maintainer, Institut d’Astrophysique de Paris (IAP).
For information on contributing, refer to CONTRIBUTING.md.
If you use the SelfiSys package in your research, please cite the following paper and feel free to contact the authors for feedback, collaboration opportunities, or other inquiries.
Diagnosing Systematic Effects Using the Inferred Initial Power Spectrum Hoellinger, T. and Leclercq, F., 2024 arXiv:2412.04443 [astro-ph.CO] [ADS] [pdf]
BibTeX entry for citation:
@ARTICLE{hoellinger2024diagnosing,
author = {Hoellinger, Tristan and Leclercq, Florent},
title = "{Diagnosing Systematic Effects Using the Inferred Initial Power Spectrum}",
journal = {arXiv e-prints},
keywords = {Astrophysics - Cosmology and Nongalactic Astrophysics, Astrophysics - Instrumentation and Methods for Astrophysics},
year = 2024,
month = dec,
eid = {arXiv:2412.04443},
pages = {arXiv:2412.04443},
doi = {10.48550/arXiv.2412.04443},
archivePrefix = {arXiv},
eprint = {2412.04443},
primaryClass = {astro-ph.CO},
adsurl = {https://ui.adsabs.harvard.edu/abs/2024arXiv241204443H},
adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}The code is written in Python 3.10 and depends on the following packages:
pySELFI: Python implementation of the Simulator Expansion for Likelihood-Free Inference.Simbelmynë: A hierarchical probabilistic simulator for generating synthetic galaxy survey data.ELFI: A statistical software package for likelihood-free inference, implementing in particular Approximate Bayesian Computation (ABC) with a Population Monte Carlo (PMC) sampler.
A comprehensive list of dependencies, including version specifications to ensure reproducibility, will be provided in a yaml file, along with installation instructions, in a future release.
This software is distributed under the GPLv3 Licence. Please review the LICENSE file in the repository to understand the terms of use and ensure compliance. By downloading and using this software, you agree to the terms of the licence.