Note

Click here to download the full example code

Advanced Usage

The Cornetto class is the stateful interface: it caches KDEs and statistics so you can render multiple figures from the same dataset without recomputing. This section also shows styling options — palettes, dark theme, truth line customisation, and the pairplot comparison layout.

Shared synthetic posterior

import numpy as np
from cornetto import Cornetto, corner

rng = np.random.default_rng(0)
N = 20_000

data = {
    "mass_1": rng.normal(30, 4, N),
    "mass_2": rng.normal(25, 3, N),
    "spin":   rng.uniform(-1, 1, N),
    "dist":   rng.exponential(400, N),
}

labels = {
    "mass_1": r"$m_1\,[M_\odot]$",
    "mass_2": r"$m_2\,[M_\odot]$",
    "spin":   r"$\chi_{\mathrm{eff}}$",
    "dist":   r"$d_L\,[\mathrm{Mpc}]$",
}

Using the Cornetto class

Instantiate once, then call .plot(), .marginal(), .trace() and .trace_marginal() as many times as needed. KDEs are computed once and cached.

c = Cornetto(
    data,
    labels=labels,
    chain_labels=["GW200129"],
    truths={"mass_1": 30.0},
    stat="median_hdi",
)

fig, axes = c.plot(color="indigo")

Summary statistics

c.summary() returns a SummaryTable — renders as an HTML table in Jupyter and as plain text in the terminal.

tbl = c.summary()
print(tbl)

Out:

+--------+----------+-----------+--------+--------+-----------+--------+
| param  | chain    | median    | p16    | p84    | mean      | std    |
+--------+----------+-----------+--------+--------+-----------+--------+
| mass_1 | GW200129 | 29.98     | 26.08  | 34     | 30.02     | 3.984  |
| mass_2 | GW200129 | 25        | 21.96  | 27.98  | 25        | 3.023  |
| spin   | GW200129 | -0.007335 | -0.683 | 0.6804 | -0.001651 | 0.5783 |
| dist   | GW200129 | 271.8     | 69.5   | 732.4  | 399.3     | 405.6  |
+--------+----------+-----------+--------+--------+-----------+--------+

All named palettes

Five built-in single-chain palettes: "indigo" (default), "coral", "teal", "gold", "ink".

for color in ["indigo", "coral", "teal", "gold", "ink"]:
    fig, axes = corner(
        {k: data[k] for k in ["mass_1", "mass_2"]},
        labels={k: labels[k] for k in ["mass_1", "mass_2"]},
        chain_labels=[color],
        color=color,
        show_titles=False,
    )

Custom truth line styling

Pass kwargs_truths to change the look of truth crosshairs and markers.

fig, axes = corner(
    data,
    labels=labels,
    truths={"mass_1": 30.0, "spin": 0.0},
    chain_labels=["GW200129"],
    kwargs_truths={"ls": "-.", "lw": 1.5, "marker": "*", "markersize": 8},
)

Pairplot — lower vs. upper triangle

Cornetto.pairplot(other) overlays two chains on a single corner grid: self on the lower triangle + diagonal, other on the upper triangle.

data_B = {
    "mass_1": rng.normal(35, 3, N),
    "mass_2": rng.normal(28, 2, N),
}

a = Cornetto(
    {k: data[k] for k in ["mass_1", "mass_2"]},
    labels={k: labels[k] for k in ["mass_1", "mass_2"]},
    chain_labels=["Model A"],
)
b = Cornetto(
    data_B,
    labels={k: labels[k] for k in ["mass_1", "mass_2"]},
    chain_labels=["Model B"],
)

fig, axes = a.pairplot(b, other_color="coral")

Annotation: tension and peak markers

show_tension=True prints the overlap-integral 2-D tension in each off-diagonal panel; annotate_peaks=True marks local PDF maxima on the diagonal.

fig, axes = corner(
    {k: data[k] for k in ["mass_1", "mass_2"]},
    labels={k: labels[k] for k in ["mass_1", "mass_2"]},
    chain_labels=["GW200129"],
    annotate_peaks=True,
)

Total running time of the script: ( 0 minutes 2.694 seconds)

Download Python source code: plot_3_advanced.py

Download Jupyter notebook: plot_3_advanced.ipynb

Gallery generated by mkdocs-gallery