pygip.models.attack.Realistic

Classes

DGLEdgePredictor(*args, **kwargs)	DGL version of edge prediction module.
DGLSurrogateModel(*args, **kwargs)	DGL version of surrogate model.
RealisticAttack(dataset, attack_x_ratio, ...)	DGL-based GNN model extraction attack with updated metrics API.

class pygip.models.attack.Realistic.DGLEdgePredictor(*args: Any, **kwargs: Any)

Bases: Module

DGL version of edge prediction module.

forward(graph, features)

predict_edges(node_embeddings, node_pairs)

Predict edge existence probability for a list of node index pairs.

class pygip.models.attack.Realistic.DGLSurrogateModel(*args: Any, **kwargs: Any)

Bases: Module

DGL version of surrogate model.

forward(graph, features)

class pygip.models.attack.Realistic.RealisticAttack(dataset, attack_x_ratio, attack_a_ratio, model_path=None, hidden_dim=64, threshold_s=0.7, threshold_a=0.5)

Bases: BaseAttack

DGL-based GNN model extraction attack with updated metrics API.

_abc_impl = <_abc_data object>

_evaluate_and_update_metrics(enhanced_graph, metric, metric_comp)

Evaluate surrogate against target on the real test set and update metric containers.

add_potential_edges(candidate_edges, labeled_nodes)

Add potential edges whose predicted probability exceeds the threshold.

attack()

Execute the attack and return two JSON-like dicts: performance and computation metrics.

compute_cosine_similarity(features)

Compute cosine similarity of node features.

generate_candidate_edges(labeled_nodes, unlabeled_nodes)

Generate candidate edges based on feature cosine similarity threshold.

simulate_target_model_queries(query_nodes, error_rate=0.15)

Query the target model for labels on query_nodes and introduce a small error rate.

supported_api_types = {'dgl'}

supported_datasets = {}

train_edge_predictor(labeled_nodes, predicted_labels, epochs=100)

Train the auxiliary edge prediction model.

train_surrogate_model(enhanced_graph, labeled_nodes, predicted_labels, epochs=200)

Train the surrogate model on queried nodes and pseudo labels.