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

State-Space GPs#

`StateSpaceConjugateModel`	Joint model for a state-space (Markovian) GP: conditioning is Kalman.
`StateSpacePrior`	Prior for a state-space (Markovian) GP.
`TruncatedPeriodic`	Truncated-Fourier approximation of the periodic kernel.
`fit`	Fit a state-space posterior with Optax (gradient-descent style).
`fit_lbfgs`	Fit a state-space posterior with Optax's L-BFGS (`while_loop` driver).
`fit_scipy`	Fit a state-space posterior with SciPy's L-BFGS-B.
`state_space_mll`	Marginal log-likelihood via the square-root Kalman filter.
`to_sde`	Convert a kernel to its state-space (linear SDE) representation.