Argus
Welcome to the Argus project documentation.
Argus is a research project for the detection of nHz gravitational waves using Pulsar Timing Arrays (PTAs), leveraging a state-space representation.
It is an ongoing effort to open-source methods developed at the University of Melbourne and OzGrav. For the accompanying papers please see:
About this Documentation
This documentation provides an overview of the project, installation and usage instructions, API details, and developer notes for contributors.
Note
If you're new to Argus, start with the "Getting Started" section.
Example Usage
You can find examples of standard PTA analysis using various datasets:
- Basic parameter estimation with Mock Data Challenge
- Multi-parameter estimation with NANOGrav data
- Custom noise models
Interactive Notebooks
Interactive Jupyter notebooks are available in the notebooks/
directory of the repository.
Key Features
- State-space framework for PTA data analysis
- Bayesian parameter estimation using NUTS sampling with NumPyro
- Gravitational wave detection and characterization
- Pulsar noise modeling with flexible priors
- JAX-based implementation for high-performance computing
Etymology
The project is named Argus, after the hundred-eyed giant of Greek mythology.
In a pulsar timing array, multiple pulsars across the sky act as a network of cosmic "eyes", continuously watching for gravitational waves.
Quick Start
To get started with Argus:
- Install the package: See the Getting Started guide
- Run your first analysis: Check out the examples
- Explore the API: Browse the API Reference
- Contribute: Read our Contributing Guide
Research and Citations
If you use Argus in your research, please cite the relevant papers:
@article{kimpson2024a,
title={State-space analysis for pulsar timing arrays},
author={Kimpson, Tom and others},
journal={arXiv preprint arXiv:2409.14613},
year={2024}
}
Support
- 🐛 Bug reports: GitHub Issues
- 💬 Discussions: GitHub Discussions
- 📧 Contact: University of Melbourne NSGW