SequentialMetaRecommender¶

class baybe.recommenders.meta.sequential.SequentialMetaRecommender[source]¶

Bases: MetaRecommender

A meta recommender that uses a pre-defined sequence of recommenders.

A new recommender is taken from the sequence whenever at least one new measurement is available, until all recommenders are exhausted. More precisely, a recommender change is triggered whenever the size of the training dataset increases; the actual content of the dataset is ignored.

Note

The provided sequence of recommenders will be internally pre-collected into a list. If this is not acceptable, consider using baybe.recommenders.meta.sequential.StreamingSequentialMetaRecommender instead.

Raises:: NoRecommendersLeftError – If more recommenders are requested than there are recommenders available and mode="raise".

Public methods

`__init__`(recommenders[, mode, _step, ...])	Method generated by attrs for class SequentialMetaRecommender.
`from_dict`(dictionary)	Create an object from its dictionary representation.
`from_json`(string)	Create an object from its JSON representation.
`recommend`(searchspace[, batch_size, ...])	See `baybe.recommenders.base.RecommenderProtocol.recommend()`.
`select_recommender`(searchspace[, ...])	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

`recommenders`	A finite-length sequence of recommenders to be used.
`mode`	Defines what shall happen when the last recommender in the sequence has been consumed but additional recommender changes are triggered:
`allow_recommending_already_measured`	Deprecated! The flag has become an attribute of `baybe.recommenders.pure.base.PureRecommender`.
`allow_repeated_recommendations`	Deprecated! The flag has become an attribute of `baybe.recommenders.pure.base.PureRecommender`.

__init__(recommenders, mode: Literal['raise', 'reuse_last', 'cyclic'] = 'raise', _step: int = -1, _n_last_measurements: int = -1, *, allow_repeated_recommendations: bool | None = None, allow_recommending_already_measured: bool | None = None)¶

Method generated by attrs for class SequentialMetaRecommender.

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.

recommend(searchspace: SearchSpace, batch_size: int = 1, train_x: DataFrame | None = None, train_y: DataFrame | None = None)¶

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

Return type:: DataFrame

select_recommender(searchspace: SearchSpace, batch_size: int = 1, train_x: DataFrame | None = None, train_y: DataFrame | None = None)[source]¶

Select a pure recommender for the given experimentation context.

Parameters:

searchspace (SearchSpace) – See baybe.recommenders.meta.base.MetaRecommender.recommend().
batch_size (int) – See baybe.recommenders.meta.base.MetaRecommender.recommend().
train_x (Optional[DataFrame]) – See baybe.recommenders.meta.base.MetaRecommender.recommend().
train_y (Optional[DataFrame]) – See baybe.recommenders.meta.base.MetaRecommender.recommend().

Return type:

PureRecommender

Returns:

The selected recommender.

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.

allow_recommending_already_measured: bool¶: Deprecated! The flag has become an attribute of baybe.recommenders.pure.base.PureRecommender.

allow_repeated_recommendations: bool¶: Deprecated! The flag has become an attribute of baybe.recommenders.pure.base.PureRecommender.

mode: Literal['raise', 'reuse_last', 'cyclic']¶

Defines what shall happen when the last recommender in the sequence has been consumed but additional recommender changes are triggered:

"raise": An error is raised.

"reuse_last": The last recommender in the sequence is used indefinitely.

"cycle": The selection restarts from the beginning of the sequence.

recommenders: list[PureRecommender]¶: A finite-length sequence of recommenders to be used. For infinite-length iterables, see baybe.recommenders.meta.sequential.StreamingSequentialMetaRecommender.