StableEMRIFisher (SEF)

StableEMRIFisher is a Python package for computing stable Fisher information matrices for Extreme Mass Ratio Inspiral (EMRI) gravitational wave sources. It provides robust numerical derivatives of waveforms and returns Fisher matrices that can be used for accelerated parameter estimation analyses or population inference.

StableEMRIFisher uses the FastEMRIWaveforms package. Please refer to the documentation to know how to get started.

Key Features

• Stable Numerical Derivatives: Robust finite difference methods for parameter derivatives for adiabatic Kerr eccentric equatorial based waveform models.
• Stability Checks: Careful validation of numerical stability and convergence of the derivatives.
• GPU Acceleration: Efficient computation for both CPUs and GPUs.
• Response Function Utilises fastlisaresponse to efficiently compute the response of the LISA detector to EMRI waveforms.
• Validated: Compared against MCMC parameter estimation studies.

Installation

Development Installation (Recommended)

If you're using conda

# Create and activate environment
conda create -n sef_env python=3.12
conda activate sef_env

# Clone and install
git clone https://github.com/perturber/StableEMRIFisher.git
cd StableEMRIFisher
pip install -e .

GPU-Accelerated Installation

For GPU acceleration, install with CUDA support. Note: GPU support requires Linux x86_64 systems with NVIDIA GPUs and appropriate CUDA drivers.

Using pip:

# For CUDA 11.x (Linux x86_64 only)
pip install -e ".[cuda11x]"

# For CUDA 12.x (Linux x86_64 only)  
pip install -e ".[cuda12x]"

StableEMRIFisher with the LISA response

• First install lisaanalysistools by following the instructions here. If using GPUs, install from source.
• Second install fastlisaresponse by following the instructions here. If using GPUs, install from source.

Documentation

Full documentation is available at stableemrifisher.readthedocs.io

• Installation Guide
• Quick Start Tutorial
• API Reference
• Examples
• Validation Studies

Building Documentation Locally

# Install documentation dependencies
pip install -e ".[docs]"

# Build documentation
cd docs
make clean
make html

# View documentation
open _build/html/index.html  # macOS
# or 
xdg-open _build/html/index.html  # Linux

Contributing

We welcome contributions! Please:

1. Report bugs via GitHub Issues
2. Suggest features through issue discussions
3. Submit pull requests with improvements
4. Improve documentation and examples

Citation

If you use StableEMRIFisher in your research, please cite it using the metadata in our CITATION.cff file.

You can easily copy the citation in APA or BibTeX format by clicking the "Cite this repository" button in the right sidebar of this GitHub page. Alternatively, you can use the following BibTeX entry:

@software{Kejriwal_StableEMRIFisher_2026,
  author       = {Kejriwal, Shubham and Burke, Ollie and Chapman-Bird, Christian and J. K. Chua, Alvin},
  title        = {StableEMRIFisher},
  version      = {2.0.0},
  year         = {2026},
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.18778938},
  url          = {[https://github.com/perturber/StableEMRIFisher](https://github.com/perturber/StableEMRIFisher)}
}

Please also cite FastEMRIWaveforms and other dependencies as appropriate.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Support

• Questions: Open a GitHub Discussion
• Bug Reports: Submit a GitHub Issue
• Documentation: Visit stableemrifisher.readthedocs.io

Acknowledgments

Development supported by:

• Centre National de la Recherche Scientifique (CNRS)
• National University of Singapore

Special thanks to the FastEMRIWaveforms development team and the LISA Consortium.