torch_geometric.datasets.NELL

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

Bases: InMemoryDataset

The NELL dataset, a knowledge graph from the “Toward an Architecture for Never-Ending Language Learning” paper. The dataset is processed as in the “Revisiting Semi-Supervised Learning with Graph Embeddings” paper.

Note

Entity nodes are described by sparse feature vectors of type torch_sparse.SparseTensor, which can be either used directly, or can be converted via data.x.to_dense(), data.x.to_scipy() or data.x.to_torch_sparse_coo_tensor().

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)

STATS:

#nodes

#edges

#features

#classes

65,755

251,550

61,278

186