torch_geometric.datasets.WordNet18

class WordNet18(root: str, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None)[source]

Bases: InMemoryDataset

The WordNet18 dataset from the “Translating Embeddings for Modeling Multi-Relational Data” paper, containing 40,943 entities, 18 relations and 151,442 fact triplets, e.g., furniture includes bed.

Note

The original WordNet18 dataset suffers from test leakage, i.e. more than 80% of test triplets can be found in the training set with another relation type. Therefore, it should not be used for research evaluation anymore. We recommend to use its cleaned version WordNet18RR instead.

Parameters

  • root (str) – Root directory where the dataset should be saved.

  • transform (callable, optional) – A function/transform that takes in an torch_geometric.data.Data object and returns a transformed version. The data object will be transformed before every access. (default: None)

  • pre_transform (callable, optional) – A function/transform that takes in an torch_geometric.data.Data object and returns a transformed version. The data object will be transformed before being saved to disk. (default: None)