diagnostics

A collection of plotting utilities and metrics for evaluating trained Workflows.

Functions

bootstrap_comparison(observed_samples, ...)	Computes the distance between observed and reference samples and generates a distribution of null sample distances by bootstrapping for hypothesis testing.
calibration_ecdf(estimates, targets[, ...])	Creates the empirical CDFs for each marginal rank distribution and plots it against a uniform ECDF.
calibration_ecdf_from_quantiles(estimates, ...)	Creates the empirical CDFs for each marginal rank distribution and plots it against a uniform ECDF.
calibration_error(estimates, targets[, ...])	Computes an aggregate score for the marginal calibration error over an ensemble of approximate posteriors.
calibration_histogram(estimates, targets[, ...])	Creates and plots publication-ready histograms of rank statistics for simulation-based calibration (SBC) checks according to [1].
loss(history[, train_key, val_key, ...])	A generic helper function to plot the losses of a series of training epochs and runs.
mc_calibration(pred_models, true_models[, ...])	Plots the calibration curves, the ECEs and the marginal histograms of predicted posterior model probabilities for a model comparison problem.
mc_confusion_matrix(pred_models, true_models)	Plots a confusion matrix for validating a neural network trained for Bayesian model comparison.
mmd_hypothesis_test(mmd_null[, ...])
pairs_posterior(estimates[, targets, ...])	Generates a bivariate pair plot given posterior draws and optional prior or prior draws.
pairs_samples([samples, variable_keys, ...])	A more flexible pair plot function for multiple distributions based upon collected samples.
posterior_contraction(estimates, targets[, ...])	Computes the posterior contraction (PC) from prior to posterior for the given samples.
recovery(estimates, targets[, ...])	Creates and plots publication-ready recovery plot with true estimate vs.
recovery_from_estimates(estimates, targets, ...)	Creates and plots publication-ready recovery plot of estimates vs.
summary_space_comparison(observed_data, ...)	Computes the distance between observed and reference data in the summary space and generates a distribution of distance values under the null hypothesis to assess model misspecification.
z_score_contraction(estimates, targets[, ...])	Implements a graphical check for global model sensitivity by plotting the posterior z-score over the posterior contraction for each set of posterior samples in estimates according to [1].