Visualize posterior samples — pairplot • neuralsbi

A pair plot built on GGally::ggpairs(): 1-D marginal densities on the diagonal and 2-D highest-density regions (via ggdensity::geom_hdr()) in the lower triangle, with optional markers for a reference (e.g. true) parameter value. Analogous to Python sbi's pairplot.

Usage

pairplot(
  samples,
  truth = NULL,
  labels = NULL,
  limits = NULL,
  col = "steelblue",
  alpha = 0.4,
  ...
)

Arguments

samples: A matrix of posterior draws (rows = draws), or an nsbi_samples object.
truth: Optional reference parameter vector to overlay.
labels: Optional parameter labels. Defaults to colnames(samples) (set automatically for sample() draws from a fit with named parameters – see prior_uniform()/prior_normal()) or theta[1], theta[2], .... Labels that parse as R syntax ("beta[1]", "rho") render as their plotmath symbol.
limits: Optional list (one c(lo, hi) per parameter, in column order) or matrix of per-parameter axis limits. Defaults to each parameter's own data range (padded 5%), applied consistently across every panel that plots it – without this, the lower-triangle and diagonal panels for the same parameter can each draw a different range, since ggdensity::geom_hdr() and ggplot2::geom_density() estimate their density grid independently per panel.
col: Density-region and marginal-density fill colour.
alpha: Marginal-density fill transparency. The lower-triangle highest-density regions shade themselves by probability level (99/95/80/50%) instead, via ggdensity's own alpha mapping.
...: Passed to the lower-triangle ggdensity::geom_hdr() layer.

Value

A ggmatrix object (also drawn as a side effect), invisibly.