torch_geometric.datasets.StochasticBlockModelDataset

class StochasticBlockModelDataset(root: str, block_sizes: Union[List[int], Tensor], edge_probs: Union[List[List[float]], Tensor], num_graphs: int = 1, num_channels: Optional[int] = None, is_undirected: bool = True, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, force_reload: bool = False, **kwargs: Any)[source]

Bases: InMemoryDataset

一个由随机块模型生成的合成图数据集。 每个块的节点特征是从正态分布中采样的，其中 簇的中心是超立方体的顶点，如 sklearn.datasets.make_classification() 方法所计算的。

Parameters:

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

  • block_sizes ([int] 或 LongTensor) – 块的大小。

  • edge_probs ([[float]] or FloatTensor) – 从一个块到另一个块的边的密度。如果图是无向的，则必须是对称的。

  • num_graphs (int, optional) – The number of graphs. (default: 1)

  • num_channels (int, optional) – The number of node features. If given as None, node features are not generated. (default: None)

  • is_undirected (bool, optional) – Whether the graph to generate is undirected. (default: True)

  • 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) – 一个函数/转换，它接收一个 torch_geometric.data.Data 对象并返回一个转换后的版本。数据对象将在保存到磁盘之前进行转换。（默认值：None）

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

  • **kwargs (可选) – 传递给sklearn.datasets.make_classification()方法的关键字参数，用于生成节点特征。