hyperoptax.acquisition#

Classes

`BaseAcquisition`()	Base class for acquisition functions.
`EI`([xi])	Expected Improvement acquisition function.
`UCB`([kappa])	Upper Confidence Bound acquisition function.

class hyperoptax.acquisition.BaseAcquisition[source]#

Base class for acquisition functions.

get_argmax(mean, std, seen_mask, n_points=1)[source]#

Parameters:

class hyperoptax.acquisition.UCB(kappa=2.0)[source]#

Upper Confidence Bound acquisition function.

__init__(kappa=2.0)[source]#

class hyperoptax.acquisition.EI(xi=0.01)[source]#

Expected Improvement acquisition function.

__init__(xi=0.01)[source]#

class hyperoptax.acquisition.PI(xi=0.01)[source]#

Probability of Improvement acquisition function.

__init__(xi=0.01)[source]#

class hyperoptax.acquisition.BaseHallucination[source]#

Base class for Kriging Believer hallucination strategies.

Any callable with signature (mean, std, key, y_max) -> scalar can be used as a hallucination strategy — subclassing is optional.

class hyperoptax.acquisition.MeanHallucination[source]#: Classical Kriging Believer: hallucinate with GP posterior mean.

class hyperoptax.acquisition.SampleHallucination[source]#

Randomized Kriging Believer (RKB): hallucinate with a posterior sample.

arXiv 2603.01470.

class hyperoptax.acquisition.UCBHallucination(kappa=2.0)[source]#

Optimistic hallucination: mean + kappa * std.

__init__(kappa=2.0)[source]#

class hyperoptax.acquisition.ConstantHallucination(value=None)[source]#

Ginsbourger et al. 2010: hallucinate with y_max or a fixed constant.

If value is None, uses the current best observed value (y_max). Otherwise uses the fixed value regardless of observations.

__init__(value=None)[source]#

hyperoptax.acquisition