Welcome to GNN-AID documentation!

GNN-AID is an open instrument for Analysis, Interpretation, and Defensing Graph Neural Networks for explaining your graph neural network with a number of existing explainers.

Check out the Usage section for further information, including how to Installation the project.

Note

This project is under active development.

Contents

• Usage
  • Installation
• API reference
  • Working with datasets
    • CustomDataset
      • CustomDataset.__init__()
      • CustomDataset.node_attributes_dir
      • CustomDataset.edge_attributes_dir
      • CustomDataset.labels_dir
      • CustomDataset.edges_path
      • CustomDataset.edge_index_path
      • CustomDataset.build()
  • Creating and managing models
    • GNNModelManager
      • GNNModelManager.__init__()
      • GNNModelManager.train_1_step()
      • GNNModelManager.train_full()
      • GNNModelManager.load_model()
      • GNNModelManager.save_model()
      • GNNModelManager.load_model_executor()
      • GNNModelManager.get_hash()
      • GNNModelManager.save_model_executor()
      • GNNModelManager.from_model_path()
      • GNNModelManager.get_full_info()
      • GNNModelManager.get_model_data()
    • FrameworkGNNModelManager
      • FrameworkGNNModelManager.additional_config
      • FrameworkGNNModelManager.__init__()
      • FrameworkGNNModelManager.init()
      • FrameworkGNNModelManager.train_1_step()
      • FrameworkGNNModelManager.train_1_step_single()
      • FrameworkGNNModelManager.train_1_step_mul()
      • FrameworkGNNModelManager.load_model()
      • FrameworkGNNModelManager.save_model()
      • FrameworkGNNModelManager.train_model()
      • FrameworkGNNModelManager.run_model()
      • FrameworkGNNModelManager.evaluate_model()
      • FrameworkGNNModelManager.compute_stats_data()
      • FrameworkGNNModelManager.send_data()
      • FrameworkGNNModelManager.send_epoch_results()
    • ProtGNNModelManager
      • ProtGNNModelManager.additional_config
      • ProtGNNModelManager.save_model()
      • ProtGNNModelManager.load_model()
      • ProtGNNModelManager.evaluate_model()
      • ProtGNNModelManager.train_full()
      • ProtGNNModelManager.run_model()
  • Interpretation
    • ProgressBar
      • ProgressBar.__init__()
      • ProgressBar.reset()
      • ProgressBar.update()
    • Explainer
      • Explainer.check_availability()
      • Explainer.__init__()
      • Explainer.save()
    • DummyExplainer
      • DummyExplainer.check_availability()
      • DummyExplainer.__init__()
  • Auxiliary functions
    • Config
      • Config.__init__()
      • Config.merge()
      • Config.to_saveable_dict()
    • DatasetConfig
      • DatasetConfig.__init__()
      • DatasetConfig.full_name()
      • DatasetConfig.from_full_name()
    • DatasetVarConfig
    • ModelStructureConfig
    • ModelConfig
      • ModelConfig.__init__()
    • ModelManagerConfig
    • ModelModificationConfig
      • ModelModificationConfig.__init__()
    • ExplainerInitConfig
      • ExplainerInitConfig.__init__()
    • ExplainerRunConfig
      • ExplainerRunConfig.__init__()
    • ExplainerModificationConfig
      • ExplainerModificationConfig.__init__()