dgl.dataloading.DistNodeDataLoader

class dgl.dataloading.DistNodeDataLoader(g, nids, graph_sampler, device=None, **kwargs)

Sampled graph data loader over nodes for distributed graph storage.

It wraps an iterable over a set of nodes, generating the list of message flow graphs (MFGs) as computation dependency of the said minibatch, on a distributed graph.

All the arguments have the same meaning as the single-machine counterpart dgl.dataloading.DataLoader except the first argument g which must be a dgl.distributed.DistGraph.

Parameters

• g (DistGraph) – The distributed graph.

• graph_sampler, device, kwargs (nids,) – See dgl.dataloading.DataLoader.

See also

dgl.dataloading.DataLoader

__init__(g, nids, graph_sampler, device=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(g, nids, graph_sampler[, device])

Initialize self.