torch_geometric.datasets.Twitch

class Twitch(root: str, name: str, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, force_reload: bool = False)[source]

Bases: InMemoryDataset

The Twitch Gamer networks introduced in the “Multi-scale Attributed Node Embedding” paper. Nodes represent gamers on Twitch and edges are followerships between them. Node features represent embeddings of games played by the Twitch users. The task is to predict whether a user streams mature content.

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

  • name (str) – The name of the dataset ("DE", "EN", "ES", "FR", "PT", "RU").

  • 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)

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

STATS:

Name

#nodes

#edges

#features

#classes

DE

9,498

315,774

128

2

EN

7,126

77,774

128

2

ES

4,648

123,412

128

2

FR

6,551

231,883

128

2

PT

1,912

64,510

128

2

RU

4,385

78,993

128

2