tapas.threat_models.attacker_knowledge.AttackerKnowledgeOnData

class tapas.threat_models.attacker_knowledge.AttackerKnowledgeOnData

Bases: object

Abstract base class that represents the knowledge that attackers have on real datasets, in the form of a prior from which they can sample training datasets for the attack. This class also requires a way to generate “testing” datasets from (potentially) another distribution, which will be used to test an attack trained on “training” datasets.

Note that these methods generate _real_ datasets – a generator still needs to be applied to produce synthetic datasets.

__init__()

Methods

__init__()
generate_datasets(num_samples[, training])	Generate num_samples training or testing datasets.

Attributes

label

A string to represent this knowledge.

abstract generate_datasets(num_samples: int, training: bool = True) → list[Dataset]

Generate num_samples training or testing datasets.

property label

A string to represent this knowledge.