Smoke render: optimisation and MCMC are shrunk for CI speed, so figures and diagnostics are not publication-fidelity. The production docs are built with full-length inference.

GPJax

HeteroscedasticPosterior#

class gpjax.gps.HeteroscedasticPosterior(prior, likelihood, jitter=1e-06)[source]#

Bases: LatentPosterior[P, HL]

Posterior shell for heteroscedastic likelihoods.

The posterior retains both the signal and noise priors; inference is delegated to variational families and specialised objectives.

Parameters:

prior (AbstractPrior)
likelihood (Any)
jitter (float)