tapas.attacks.distances.HammingDistance

class tapas.attacks.distances.HammingDistance(columns: Optional[list[str]] = None)

Bases: tapas.attacks.distances.DistanceMetric

Hamming distance (“L_0”): counts the number of attributes that are identical between two records. While this is mainly for categorical attributes, it also works with continuous values.

__init__(columns: Optional[list[str]] = None)

Parameters

columns (list of column names, optional (None)) – List of the columns on which to compute the distance. If this is not provided, all columns are used. Only for tabular datasets.

Methods

__init__([columns])

param columns List of the columns on which to compute the distance. If this is

Attributes

label