How to organize data when I want to use multi-GPU in one server.

[Zrealshadow]

Hi,
I want to train the model on multi-GPU on same server.
Currently, I use NeighborLoader to generate sub-HeteroData as a mini-batch for model training.
I notice there is a DataListLoader supporting multi-GPU. but is there any class like NeighborListLoader ?

As for the tutorial, I notice the tutorial in https://pytorch-geometric.readthedocs.io/en/2.6.0/tutorial/distributed_pyg.html
which is more related to distributed training.

Is there any tutorial for multi-GPU training on same server ?

[Zrealshadow]

like this example:
https://github.com/pyg-team/pytorch_geometric/blob/master/examples/multi_gpu/data_parallel.py