torch_geometric.datasets.AirfRANS

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

Bases: InMemoryDataset

The AirfRANS dataset from the “AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier-Stokes Solutions” paper, consisting of 1,000 simulations of steady-state aerodynamics over 2D airfoils in a subsonic flight regime. The different tasks ("full", "scarce", "reynolds", "aoa") define the utilized training and test splits.

Each simulation is given as a point cloud defined as the nodes of the simulation mesh. Each point of a point cloud is described via 5 features: the inlet velocity (two components in meter per second), the distance to the airfoil (one component in meter), and the normals (two components in meter, set to 0 if the point is not on the airfoil). Each point is given a target of 4 components for the underyling regression task: the velocity (two components in meter per second), the pressure divided by the specific mass (one component in meter squared per second squared), the turbulent kinematic viscosity (one component in meter squared per second). Finaly, a boolean is attached to each point to inform if this point lies on the airfoil or not.

A library for manipulating simulations of the dataset is available here.

The dataset is released under the ODbL v1.0 License.

Note

Data objects contain no edge indices to be agnostic to the simulation mesh. You are free to build a graph via the torch_geometric.transforms.RadiusGraph transform.

Parameters:

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

  • task (str) – The task to study ("full", "scarce", "reynolds", "aoa") that defines the utilized training and test splits.

  • train (bool, optional) – If True, loads the training dataset, otherwise the test dataset. (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) – 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)

STATS:

#graphs

#nodes

#edges

#features

#tasks

1,000

~180,000

0

5

4