pygip.models.defense.BackdoorWM¶

Module Attributes

class pygip.models.defense.BackdoorWM.BackdoorWM(dataset, attack_node_fraction, model_path=None, trigger_rate=0.01, l=20, target_label=0)[source]¶

Bases: BaseDefense

_abc_impl = <_abc_data object>¶

_load_model()[source]¶: Load a pre-trained model.

defend()[source]¶: Execute the backdoor watermark attack.

evaluate_model(model, features, labels)[source]¶: Evaluate model performance

inject_backdoor_trigger(data, trigger_rate=None, trigger_feat_val=0.99, l=None, target_label=None)[source]¶: Feature-based Trigger Injection

supported_api_types = {'dgl'}¶

train_target_model()[source]¶: Train the target model with backdoor injection.

verify_backdoor(model, trigger_nodes, target_label)[source]¶: Verify backdoor attack success rate