torch_geometric.datasets.PCPNetDataset

class PCPNetDataset(root: str, category: str, split: str = 'train', transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None, force_reload: bool = False)[source]

Bases: InMemoryDataset

The PCPNet dataset from the “PCPNet: Learning Local Shape Properties from Raw Point Clouds” paper, consisting of 30 shapes, each given as a point cloud, densely sampled with 100k points. For each shape, surface normals and local curvatures are given as node features.

Parameters:

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

  • category (str) – The training set category (one of "NoNoise", "Noisy", "VarDensity", "NoisyAndVarDensity" for split="train" or split="val", or one of "All", "LowNoise", "MedNoise", "HighNoise", :obj:”VarDensityStriped”, "VarDensityGradient" for split="test").

  • split (str, optional) – If "train", loads the training dataset. If "val", loads the validation dataset. If "test", loads the test dataset. (default: "train")

  • 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)

  • pre_filter (callable, optional) – A function that takes in an torch_geometric.data.Data object and returns a boolean value, indicating whether the data object should be included in the final dataset. (default: None)

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