CompositeSurrogate

class baybe.surrogates.composite.CompositeSurrogate

Bases: SerialMixin, SurrogateProtocol

A class for composing multi-target surrogates from single-target surrogates.

Public methods

__init__(surrogates)	Method generated by attrs for class CompositeSurrogate.
fit(searchspace, objective, measurements)	Fit the surrogate to training data in a given modelling context.
from_dict(dictionary)	Create an object from its dictionary representation.
from_json(string)	Create an object from its JSON representation.
from_replication(surrogate)	Replicate a given single-target surrogate logic for multiple targets.
posterior(candidates)	Compute the posterior for candidates in experimental representation.
posterior_stats(candidates[, stats])	See baybe.surrogates.base.Surrogate.posterior_stats().
to_botorch()	Create the botorch-ready representation of the fitted model.
to_dict()	Create an object's dictionary representation.
to_json()	Create an object's JSON representation.

Public attributes and properties

surrogates

An index-based mapping from target names to single-target surrogates.

__init__(surrogates: _SurrogateGetter)

Method generated by attrs for class CompositeSurrogate.

For details on the parameters, see Public attributes and properties.

fit(searchspace: SearchSpace, objective: Objective, measurements: DataFrame)

Fit the surrogate to training data in a given modelling context.

For details on the expected method arguments, see baybe.recommenders.base.RecommenderProtocol().

Return type:

None

classmethod from_dict(dictionary: dict)

Create an object from its dictionary representation.

Parameters:

dictionary (dict) – The dictionary representation.

Return type:

TypeVar(_T)

Returns:

The reconstructed object.

classmethod from_json(string: str)

Create an object from its JSON representation.

Parameters:

string (str) – The JSON representation of the object.

Return type:

TypeVar(_T)

Returns:

The reconstructed object.

classmethod from_replication(surrogate: SurrogateProtocol)

Replicate a given single-target surrogate logic for multiple targets.

Return type:

CompositeSurrogate

posterior(candidates: pd.DataFrame)

Compute the posterior for candidates in experimental representation.

The (independent joint) posterior is represented as a collection of individual posterior models computed per target of the involved objective. For details, see baybe.surrogates.base.Surrogate.posterior().

Return type:

PosteriorList

posterior_stats(candidates: DataFrame, stats: Sequence[float | Literal['mean', 'std', 'var', 'mode']] = ('mean', 'std'))

See baybe.surrogates.base.Surrogate.posterior_stats().

Return type:

DataFrame

to_botorch()

Create the botorch-ready representation of the fitted model.

The botorch.models.model.Model created by this method needs to be configured such that it can be called with candidate points in computational representation, that is, input of the form as obtained via baybe.searchspace.core.SearchSpace.transform().

Return type:

ModelList

to_dict()

Create an object’s dictionary representation.

Return type:

dict

to_json()

Create an object’s JSON representation.

Return type:

str

Returns:

The JSON representation as a string.

surrogates: _SurrogateGetter

An index-based mapping from target names to single-target surrogates.