egc.utils.data_loader package

Submodules

egc.utils.data_loader.naive_data_loader module

Naive Graph Batch Data Loader

class egc.utils.data_loader.naive_data_loader.NaiveDataLoader(graph: DGLGraph, batch_size: int = 1024, n_layers: int = 1, fanouts: List[int] = -1, aug_types: List | None = None, device: device = device(type='cpu'), drop_last: bool = False)[source]

Bases: object

Naive DataLoader using full neighbor sampler

Parameters:

  • graph (dgl.DGLGraph) – graph.

  • batch_size (int, optional) – batch size. Defaults to 1024.

  • n_layers (int, optional) – GNN layers. Defaults to 1.

  • fanout (List[int] or int, optional) –

    List of neighbors to sample for each GNN layer, with the i-th element being the fanout for the i-th GNN layer.Defaults to -1.

    • If only a single integer is provided, DGL assumes that every layer will have the same fanout.

    • If -1 is provided on one layer, then all inbound edges will be included.

  • aug_types (List, optional) – augmentation types list. Defaults to [“none”].

  • drop_last (bool, optional) – set to True to drop the last incomplete batch. Defaults to False.

Raises:

ValueError – raise if exists any augmentation type not supported

get_aug_graphs() → List[DGLGraph][source]

Get the augmented graphs.

Returns:

list of graphs augmented.

Return type:

List[dgl.DGLGraph]

Module contents

Data Loaders