Zero

class gpjax.mean_functions.Zero

Bases: Constant

Zero mean function.

The zero mean function. This function returns a zero scalar value for all inputs. Unlike the Constant mean function, the constant scalar zero is fixed, and cannot be treated as a model hyperparameter and learned during training.

The constant is wrapped in paramax.non_trainable to enforce this: fit treats every array leaf as trainable, so a bare array would be optimised away from zero.

Expand for references to gpjax.mean_functions.Zero

GPJax / “Hello, GP!”

Migrations / 0.18.x → 1.0.0 / The conditioning API

ConjugateModel

Prior

StateSpaceConjugateModel

StateSpacePrior

fit

fit_lbfgs

fit_scipy

state_space_mll