molprop.data_processing.PyGDataModule

class molprop.data_processing.PyGDataModule(dataset, config, val_dataset=None, test_dataset=None, shuffle=True, split=True)

Bases: ABC

PyG DataModule that splits the dataset into three sets train, val and test.

Parameters:

• dataset (torch.utils.data.Dataset) – Dataset to be splitted.

• config – Contains the parameters defined in defaultconfig.yaml.

get_joint_dataloader(use_traindata=True, use_valdata=True, use_testdata=True, unique_index=False)

Extracts data from the initial datamodule for usage in explainability.

update_config_w_data_features(config)

Adds num node and edge features to the config.

Parameters:

config (Config) – Config file that is updated.

Returns:

Updated config file.

Return type:

Config