tabensemb.data.AbstractSplitter

class tabensemb.data.AbstractSplitter(train_val_test: List | ndarray | None = None, cv: int = -1)

Bases: object

The base class for data-splitters that split the dataset and return training, validation, and testing indices.

Attributes:

support_cv

Whether the _next_cv() is implemented.

Methods

__init__(train_val_test: List | ndarray | None = None, cv: int = -1)

reset_cv([cv])
split(df, cont_feature_names, ...[, cv])	Split the dataset.
_check_exist(df, arg, name)
_check_split(train_indices, val_indices, ...)	Check whether split indices overlap with each other.
_next_cv(df, cont_feature_names, ...)	Get the next fold of indices of training, validation, and testing sets.
_sklearn_k_fold(data, cv)	Generate a sklearn.model_selection.KFold instance and return its __next__() result.
_split(df, cont_feature_names, ...)