humancompatible.detect.binarizer.Binarizer module

class humancompatible.detect.binarizer.Binarizer.Bin(feature: Feature, operation: Operation, value: float | tuple[float, float] | int | str | list[int | str] | bool)[source]

Bases: object

evaluate(values: ndarray[int | float | str]) → ndarray[bool][source]

negate_self()[source]

class humancompatible.detect.binarizer.Binarizer.Binarizer(data_handler: DataHandler, target_positive_vals: list[int | str] | None = None)[source]

Bases: object

Handles binarizing the dataset

property data_handler

encode(X: ndarray | DataFrame, include_negations=False, include_binary_negations=False) → ndarray[bool][source]

encode_y(y: ndarray | Series) → ndarray[bool][source]

feature_names(include_negations=False, include_binary_negations=False) → list[str][source]

get_bin_encodings(include_negations=False, include_binary_negations=False, return_flat=True)[source]

multi_index_feats(include_negations=False, include_binary_negations=False) → MultiIndex[source]

target_name() → tuple[str, str][source]

class humancompatible.detect.binarizer.Binarizer.Operation(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)[source]

Bases: Enum

BETWEEN = 'between'

EQ = '='

GE = '>='

GT = '>'

IN = 'in'

LE = '<='

LT = '<'

NE = '!='

NOT_IN = 'not in'

OUTSIDE = 'outside'

classmethod negated(op) → Operation[source]

classmethod perform(op: Operation, vals: ndarray[int | float | str], reference: float | tuple[float, float] | int | str | list[int | str] | bool) → ndarray[bool][source]