BrushEstimator

class pybrush.BrushEstimator.BrushEstimator(**kwargs)

This is the base class for Brush estimators using the c++ engine.

Parameters are defined and documented in EstimatorInterface

Attributes

best_estimator_pybrush.Program

The final model picked from training. Used in subsequent calls to predict().

archive_list[deap_api.DeapIndividual]

The final population from training.

data_pybrush.Dataset

The complete data in Brush format.

train_pybrush.Dataset

Partition of data_ containing `(1-validation_size)`% of the data, in Brush format.

validation_pybrush.Dataset

Partition of data_ containing `(validation_size)`% of the data, in Brush format.

search_space_a Brush SearchSpace object.

Holds the operators and terminals and sampling utilities to update programs.

> NOTE: as for now, when serializing the model with pickle, the objects of type Dataset and SearchSpace are not serialized.

fit(X, y)

Fit an estimator to X,y.

Parameters

Xnp.ndarray

2-d array of input data.

ynp.ndarray

1-d array of (boolean) target values.

get_params(deep=True)

Get parameters for this estimator.

Parameters

deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

paramsdict

Parameter names mapped to their values.

partial_fit(X, y, *, lock_nodes_depth=0, keep_leaves_unlocked=True, keep_current_weights=False)

Fit an estimator to X,y, without reseting the estimator.

Parameters

Xnp.ndarray

2-d array of input data.

ynp.ndarray

1-d array of (boolean) target values.

lock_nodes_depthint, optional

The depth of the tree to lock. Default is 0.

keep_leaves_unlockedbool, optional

Whether to skip leaves when locking nodes. Default is True.

keep_current_weightsbool, optional

Whether to keep current weights at the spot they appear, and preventing them to be changed during optimization. Default is False.

predict(X)

Predict using the best estimator in the archive.

set_partial_fit_request(*, keep_current_weights: bool | None | str = '$UNCHANGED$', keep_leaves_unlocked: bool | None | str = '$UNCHANGED$', lock_nodes_depth: bool | None | str = '$UNCHANGED$') → BrushEstimator

Configure whether metadata should be requested to be passed to the partial_fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to partial_fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to partial_fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters

keep_current_weightsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for keep_current_weights parameter in partial_fit.

keep_leaves_unlockedstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for keep_leaves_unlocked parameter in partial_fit.

lock_nodes_depthstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for lock_nodes_depth parameter in partial_fit.

Returns

selfobject

The updated object.