BPt.CVStrategy#

class BPt.CVStrategy(groups=None, stratify=None, train_only_subjects=None)[source]#

This objects is used to encapsulate a set of parameters representing a cross-validation strategy.

Parameters

groupsstr or None, optional

The str should refer to the column key in which to preserve groups by during any CV splits. To create a combination of unique values, use Dataset.add_unique_overlap().

Note: the passed column must be of type category as well.

default = None

stratifystr or None, optional

The str input should refer to a loaded non input variable which if of type category. It will assign it as a value to preserve distribution of groups by during any during any CV splits.

To create a combination of unique values, use Dataset.add_unique_overlap().

Note: the passed column must be of type category as well.

Any target_cols passed must be categorical or binary, and cannot be float. Though you may consider creating a binary / k bin copy of a float / cont. type target variable.

default = None

train_only_subjectsNone or Subjects, optional

If passed any valid Subjects style input here, these subjects will be condiered train only and will be assigned to every training fold, and never to a testing or validation fold.

default = None

Methods

`copy`()	This method returns a deepcopy of the base object.
`get_params`([deep])	Get parameters for this estimator.
`set_params`(**params)	Set the parameters of this estimator.