Installation

SIGWAY is a pure-Python package built on JAX.

Requirements

• Python ≥ 3.11
• jax, diffrax, numpy, scipy, matplotlib (pulled in automatically)

Install

From PyPI:

pip install sigway

To work on the code, install the development version from a clone in editable mode:

git clone https://github.com/jonaselgammal/SIGWAY.git
cd SIGWAY
pip install -e .

Editable (-e) means changes to the source take effect without reinstalling.

Use float64

SIGW integrals span many orders of magnitude. Enable double precision before any other JAX call, at the top of your script or notebook:

import jax
jax.config.update("jax_enable_x64", True)

Check it worked

import jax; jax.config.update("jax_enable_x64", True)
import jax.numpy as jnp
from sigway.spectrum import OmegaGW
from sigway.kernels import RadiationKernel
from sigway.perturbations import AnalyticPerturbations

model = OmegaGW(
    AnalyticPerturbations(lambda k, A: A * jnp.ones_like(k), ("A",)),
    RadiationKernel(),
    s=jnp.linspace(0.0, 1.0, 10),
    t=jnp.geomspace(1e-4, 1e2, 200),
)
print(model.parameter_names)        # ('A',)

If that prints ('A',) you're ready for the simple example.