torch_geometric.datasets.MovieLens

class MovieLens(root: str, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, model_name: Optional[str] = 'all-MiniLM-L6-v2', force_reload: bool = False)[source]

Bases: InMemoryDataset

A heterogeneous rating dataset, assembled by GroupLens Research from the MovieLens web site, consisting of nodes of type "movie" and "user". User ratings for movies are available as ground truth labels for the edges between the users and the movies ("user", "rates", "movie").

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

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

  • model_name (str) – Name of model used to transform movie titles to node features. The model comes from the`Huggingface SentenceTransformer <https://huggingface.co/sentence-transformers>`_.

  • force_reload (bool, optional) – Whether to re-process the dataset. (default: False)