Quick Start¤

Basic Usage¤

To generate a gravitational-wave waveform, instantiate the model class and call it with a frequency array and a parameter dictionary:

import jax.numpy as jnp
import ripplegw

# Frequency grid: 20–1024 Hz at 0.25 Hz resolution
frequency = jnp.arange(20.0, 1024.0, 0.25)

# GW150914-like binary black hole parameters
params = {
    "M_c": 28.3,   # chirp mass [solar masses]
    "eta": 0.247,  # symmetric mass ratio
    "s1_z": 0.0,   # primary aligned spin
    "s2_z": 0.0,   # secondary aligned spin
    "d_L": 440.0,  # luminosity distance [Mpc]
    "phase_c": 0.0,
    "iota": 0.0,
}

# Instantiate the waveform model
waveform = ripplegw.IMRPhenomD(f_ref=20.0)

# Evaluate: returns a dict with keys "p" (h+) and "c" (hx)
polarizations = waveform(frequency, params)
hp = polarizations["p"]
hc = polarizations["c"]

All waveform models share the same interface, so switching models only requires changing one line:

waveform = ripplegw.IMRPhenomXAS(f_ref=20.0)   # same params dict
waveform = ripplegw.TaylorF2(f_ref=20.0)       # add lambda_1, lambda_2 for BNS

See ripplegw.waveform_preset for the full list of available models.

GPU and Gradient Support¤

ripple waveforms are pure JAX functions, so they work out of the box with jax.jit, jax.grad, and jax.vmap:

import jax

# JIT-compile for fast repeated evaluation
fast_waveform = jax.jit(waveform)

# Compute gradient w.r.t. chirp mass
def log_likelihood(M_c):
    h = ripplegw.IMRPhenomD()(frequency, {**params, "M_c": M_c})
    return -0.5 * jnp.sum(jnp.abs(h["p"]) ** 2)

grad_Mc = jax.grad(log_likelihood)(params["M_c"])

GPU execution requires no code changes — JAX will automatically use the GPU if one is available. See the Installation page for GPU setup.