PixelPop
Package for nonparameteric (AKA weakly modeled, data-driven) Bayesian inference of a gravitational wave population, built on JAX and numpyro.
Aimed particularly at correlated nonparameteric inference in spaces with dimension 2-3.
This method works by binning the space into a cartesian grid, and inferring the log-rate density in each bin, each of which is a free parameter. Each bin is coupled to its nearest-neighbors using an intrinsic conditional-autoregressive (ICAR) model.
The dimension of the inference problem can become very large (e.g. 10^4 for a 2-dimensional space with a density of 100 bins along each axis), and
we leverage auto-differentiation and GPU acceleration in JAX, as well as the efficient No-U-Turn HMC sampler in numpyro to sample the posterior.
Running PixelPop
Please see the example run scripts in the examples/ directory.
Attribution
Please cite Heinzel et al. (2025) if you use PixelPop in your research.
@article{Heinzel:2024jlc,
author = "Heinzel, Jack and Mould, Matthew and {\'A}lvarez-L{\'o}pez, Sof{\'\i}a and Vitale, Salvatore",
title = "{High resolution nonparametric inference of gravitational-wave populations in multiple dimensions}",
eprint = "2406.16813",
archivePrefix = "arXiv",
primaryClass = "astro-ph.HE",
doi = "10.1103/PhysRevD.111.063043",
journal = "Phys. Rev. D",
volume = "111",
number = "6",
pages = "063043",
year = "2025"
}
Additionally, consider citing Heinzel et al. (2025) which applies PixelPop to GWTC-3
@article{Heinzel:2024hva,
author = "Heinzel, Jack and Mould, Matthew and Vitale, Salvatore",
title = "{Nonparametric analysis of correlations in the binary black hole population with LIGO-Virgo-KAGRA data}",
eprint = "2406.16844",
archivePrefix = "arXiv",
primaryClass = "astro-ph.HE",
doi = "10.1103/PhysRevD.111.L061305",
journal = "Phys. Rev. D",
volume = "111",
number = "6",
pages = "L061305",
year = "2025"
},
and Alvarez-Lopez et al. (2025) which shows PixelPop can accurately recover the complex, multi-dimensional correlations in a realistic population-synthesis population.
@article{Alvarez-Lopez:2025ltt,
author = "Alvarez-Lopez, Sofia and Heinzel, Jack and Mould, Matthew and Vitale, Salvatore",
title = "{Nowhere left to hide: revealing realistic gravitational-wave populations in high dimensions and high resolution with PixelPop}",
eprint = "2506.20731",
archivePrefix = "arXiv",
primaryClass = "astro-ph.HE",
month = "6",
year = "2025"
}
2 maintainers
Metadata
2 maintainers
Release files for pixelpop 0.3.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| pixelpop-0.3.4.tar.gz | 18.6 MB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| pixelpop-0.3.4-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 18.7 MB
Release files / pixelpop-0.3.4.tar.gz
| Download URL | pixelpop-0.3.4.tar.gz |
|---|---|
| Size | 18.6 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9bc59f4150e4905beae34dd5ab95d333ff6843d18411dd5464f7a9e96adb9eaf
|
|
BLAKE2b-256 checksum How to use checksums |
129a5b05b55c2b39d4d7b60e5bfff1901f952f134adfbfde93d013b825396bf5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.6
|
Release files / pixelpop-0.3.4-py3-none-any.whl
| Download URL | pixelpop-0.3.4-py3-none-any.whl |
|---|---|
| Size | 102.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b06b5402ad3df867102d11aacbcdfa475b34552a088cf5acd986a88d9e206a10
|
|
BLAKE2b-256 checksum How to use checksums |
50c1c02918f4c999a6822d7ac31c33725902e9a1a94bf3d93c16854ef6d04e69
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.6
|
Security for pixelpop 0.3.4
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.
