TwoPhaseMetaRecommender

class baybe.recommenders.meta.sequential.TwoPhaseMetaRecommender

Bases: MetaRecommender

A two-phased meta recommender that switches at a certain specified point.

The recommender is switched when a new (batch) recommendation is requested and the training data set size (i.e., the total number of collected measurements including those gathered before the meta recommender was active) is equal to or greater than the number specified via the switch_after parameter.

Note

Throughout each phase, the meta recommender reuses the same recommender object, that is, no new instances are created. Therefore, special attention is required when using the meta recommender with stateful recommenders.

Public methods

__init__([initial_recommender, recommender, ...])	Method generated by attrs for class TwoPhaseMetaRecommender.
from_dict(dictionary)	Create an object from its dictionary representation.
from_json(string)	Create an object from its JSON representation.
get_current_recommender()	Get the current recommender, if available.
get_next_recommender(batch_size, searchspace)	Get the recommender for the next recommendation.
recommend(batch_size, searchspace[, ...])	See baybe.recommenders.base.RecommenderProtocol.recommend().
select_recommender(batch_size[, ...])	Select a pure recommender for the given experimentation context.
to_dict()	Create an object's dictionary representation.
to_json()	Create an object's JSON representation.

Public attributes and properties

initial_recommender	The initial recommender used by the meta recommender.
recommender	The recommender used by the meta recommender after the switch.
switch_after	The number of experiments after which the recommender is switched for the next requested batch.

__init__(initial_recommender: PureRecommender = NOTHING, recommender: PureRecommender = NOTHING, switch_after: int = 1)

Method generated by attrs for class TwoPhaseMetaRecommender.

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

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.

get_current_recommender()

Get the current recommender, if available.

Return type:

PureRecommender

get_next_recommender(batch_size: int, searchspace: SearchSpace, objective: Objective | None = None, measurements: DataFrame | None = None, pending_experiments: DataFrame | None = None)

Get the recommender for the next recommendation.

Returns the next recommender in row that has not yet been used for generating recommendations. In case of multiple consecutive calls, this means that the same recommender instance is returned until its recommend() method is called.

See baybe.recommenders.base.RecommenderProtocol.recommend() for details on the method arguments.

Return type:

PureRecommender

recommend(batch_size: int, searchspace: SearchSpace, objective: Objective | None = None, measurements: DataFrame | None = None, pending_experiments: DataFrame | None = None)

See baybe.recommenders.base.RecommenderProtocol.recommend().

Return type:

DataFrame

select_recommender(batch_size: int, searchspace: SearchSpace | None = None, objective: Objective | None = None, measurements: DataFrame | None = None, pending_experiments: DataFrame | None = None)

Select a pure recommender for the given experimentation context.

See baybe.recommenders.base.RecommenderProtocol.recommend() for details on the method arguments.

Return type:

PureRecommender

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.

initial_recommender: PureRecommender

The initial recommender used by the meta recommender.

recommender: PureRecommender

The recommender used by the meta recommender after the switch.

switch_after: int

The number of experiments after which the recommender is switched for the next requested batch.