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

ChainedPosterior#

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

Bases: HeteroscedasticPosterior[P, HL]

Posterior routed for heteroscedastic likelihoods using chained bounds.

Parameters:

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