torch_geometric.datasets.BAMultiShapesDataset

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

Bases: InMemoryDataset

用于评估可解释性算法的合成BA-Multi-Shapes图分类数据集，如“Global Explainability of GNNs via Logic Combination of Learned Concepts”论文中所述。

给定三个原子图案，即房屋（H）、轮子（W）和网格（G）， BAMultiShapesDataset 包含1000个图，每个图通过将图案附加到随机Barabasi-Albert（BA）图上获得，如下所示：

  • 类别 0: \(\emptyset \lor H \lor W \lor G \lor \{ H, W, G \}\)

  • 第一类: \((H \land W) \lor (H \land G) \lor (W \land G)\)

该数据集是从官方实现中预先计算得出的。

Parameters:

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

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

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

  • pre_filter (Optional[Callable], 默认值: None) – 一个函数，接收一个 torch_geometric.data.Data 对象并返回一个布尔值，表示该数据对象是否应包含在最终数据集中。

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

统计:

#图表

#节点

#edges

#特性

#classes

1000

40

~87.0

10

2