sfts 0.1.1

Scalable data-analysis framework for long-duration gravitational-wave signals

sfts

Short Fourier Transforms for Fresnel-weighted Template Summation.

Implementation of gravitational-wave data-analysis tools described in Tenorio & Gerosa (2025) to operate using Short Fourier Transforms (SFTs).

See this simple example for a quick-start on using iphenot (/ˈaɪv ˈnɒt/) and SFTs.

The package is composed of two main modules:

1. iphenot.py: jaxified re-implementation of the inspiral part of the IMRPhenomT waveform approximant.
2. kernels.py: Fresnel and Dirichlet kernels to compute scalar products using SFTs.

How to install

sfts can be pulled in from PyPI:

$ pip install sfts

To pull in jax's GPU capabilities, use:

$ pip install sfts[cuda]

Alternatively, this repository itself is pip-installable.

Cite

If the tools provided by sfts were useful to you, we would appreciate a citation of the accompanying paper:

@article{Tenorio:2025yca,
    author = "Tenorio, Rodrigo and Gerosa, Davide",
    title = "{SFTs: a scalable data-analysis framework for long-duration gravitational-wave signals}",
    eprint = "2502.11823",
    archivePrefix = "arXiv",
    primaryClass = "gr-qc",
    month = "2",
    year = "2025"
}

Whenever applicable, please consider also citing the IMRPhenomT papers listed here.