DiagnosticsPlots.prior_vs_posterior#

DiagnosticsPlots.prior_vs_posterior(var_names=None, kind='kde', idata=None, dims=None, figsize=None, backend=None, return_as_pc=False, visuals=None, aes=None, aes_by_visuals=None, **pc_kwargs)[source]#

Overlay prior and posterior 1-D marginal KDE distributions.

Thin wrapper around azp.plot_prior_posterior, which handles the prior/posterior colour legend automatically.

Parameters:
var_nameslist[str] | str | None, optional

Variable(s) to plot. None plots all variables present in both groups.

kindstr, default “kde”

Plot kind forwarded to azp.plot_prior_posterior.

idataaz.InferenceData, optional

Override instance data for this call only.

dimsdict[str, Any], optional

Coordinate filters, e.g. {"channel": ["tv"]}.

figsizetuple[float, float], optional

Figure size forwarded via figure_kwargs.

backendstr, optional

Rendering backend. Non-matplotlib backends require return_as_pc=True.

return_as_pcbool, default False

If True, return the raw PlotCollection.

visualsdict, optional

Forwarded to azp.plot_prior_posterior.

aesdict, optional

Forwarded to azp.plot_prior_posterior as an explicit keyword argument.

aes_by_visualsdict, optional

Forwarded to azp.plot_prior_posterior.

**pc_kwargs

Forwarded to azp.plot_prior_posterior.

Returns:
tuple[Figure, NDArray[Axes]] or PlotCollection

Examples

fig, axes = mmm.plot.diagnostics.prior_vs_posterior()
fig, axes = mmm.plot.diagnostics.prior_vs_posterior(
    var_names=["alpha"], dims={"channel": ["tv"]}
)