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

SigmoidBounded#

class gpjax.parameters.SigmoidBounded(value, *, low=0.0, high=1.0)[source]#

Bases: AbstractUnwrappable

Parameter bounded to [low, high] via sigmoid bijection.

Parameters:

low (float)
high (float)

unwrap()[source]#: Returns the unwrapped pytree, assuming no wrapped subnodes exist.