Quick Start¤

This page gives you a bird's-eye view of Jim's main components and how they fit together. For deeper dives into each piece, see the Tutorials.

Overview¤

A Jim analysis is assembled from the following building blocks:

Waveform Model ── Likelihood ───────────────┬──────────── 1 ──┐
                      │                     │                 │
Data ─────────────────┘          Likelihood Transforms ── 2 ──│
                                            │                 ├─→ Jim 
                    Prior ──────────────────┴──────────── 3 ──│
                      │                                       │
                      └─────────── Sample Transforms ──── 4 ──┘

Data¤

Detector data lives in Detector objects. You can fetch public LIGO/Virgo strain from GWOSC, load local files, or inject a simulated signal:

from jimgw.core.single_event.detector import get_H1, get_L1
from jimgw.core.single_event.data import Data

H1 = get_H1()
L1 = get_L1()

# Option 1: fetch from GWOSC
data = Data.from_gwosc("H1", gps_start, gps_end)
H1.set_data(data)

# Option 2: load from a .npz file
data = Data.from_file("path/to/data.npz")
H1.set_data(data)

Waveform Model¤

Jim uses ripple waveform models, which are JAX-native and fully differentiable:

from jimgw.core.single_event.waveform import RippleIMRPhenomD

waveform = RippleIMRPhenomD(f_ref=20.0)

Likelihood¤

The likelihood connects detector data with a waveform model. TransientLikelihoodFD is the standard frequency-domain likelihood for compact binary signals:

from jimgw.core.single_event.likelihood import TransientLikelihoodFD

likelihood = TransientLikelihoodFD(
    detectors=[H1, L1],
    waveform=waveform,
    trigger_time=gps_time,
    f_min=20.0,
    f_max=1024.0,
)

Prior¤

Priors are built by combining components with CombinePrior:

from jimgw.core.prior import CombinePrior, UniformPrior, SinePrior, CosinePrior

prior = CombinePrior([
    UniformPrior(10.0, 80.0, ["M_c"]),
    UniformPrior(0.125, 1.0, ["q"]),
    SinePrior(["iota"]),
    CosinePrior(["dec"]),
    # ... add more parameters
])

Transforms¤

Jim uses two kinds of transforms to bridge three parameter spaces:

     ┌───────── Likelihood Transforms ────────→ Likelihood Space
Prior Space
     └─────────── Sample Transforms ───────────→ Sampling Space

Likelihood transforms — map from the prior parameter space to the likelihood parameter space. The likelihood space is fixed by your waveform model (e.g. ripple expects eta, Cartesian spins), so the likelihood transforms you need depend on how you define your prior. For example, if your prior is on mass ratio q but the waveform expects symmetric mass ratio eta, a likelihood transform handles that conversion.
Sample transforms — map from the prior space to the sampling space. This lets the sampler explore a different parameterisation than the one your prior is defined in, typically one where correlations between parameters are reduced (e.g. sampling in detector-frame sky coordinates instead of equatorial coordinates).

Putting It Together¤

from jimgw.core.jim import Jim

jim = Jim(
    likelihood=likelihood,
    prior=prior,
    sample_transforms=sample_transforms,
    likelihood_transforms=likelihood_transforms,
    n_chains=500,
    n_training_loops=20,
    n_production_loops=10,
    # ... other hyperparameters
)

jim.sample()
samples = jim.get_samples()

For a full worked example, see the Getting Started tutorial. For production-grade scripts, browse the example/ directory on GitHub.