get_gli_dataset

gli.dataloading.get_gli_dataset(dataset: str, task: str, task_id: int = 1, device: str = 'cpu', verbose: bool = False) → DGLDataset

Get a graph dataset given dataset name and task config.

Parameters:

  • dataset (str) – graph dataset name.

  • task (str) – task type, e.g. “NodeClassification”, “NodeRegression”.

  • task_id (int, optional.) – task ID defined in dataset folder, defaults to 1.

  • device (str, optional) – device name, defaults to “cpu”.

  • verbose (bool, optional) – verbose level, defaults to False.

Return type:

dgl.data.DGLDataset.

This function essentially performs the following steps:

g = get_gli_graph(dataset, device=device, verbose=verbose) t =
get_gli_task(dataset, task, task_id=task_id, verbose=verbose) return
combine_graph_and_task(g, t)

Note

gli.dataloading.get_gli_dataset() will download the data files if the data files do not exist in the local file system.

Examples

>>> d = get_gli_dataset(dataset="cora", task="NodeClassification")
>>> d.name
'CORA dataset. NodeClassification'