pytorch-AGNN
pytorch-AGNN copied to clipboard
Pytorch implementation of the Attention-based Graph Neural Network(AGNN)
File ".../pytorch-AGNN\agnn\model.py", line 28, in forward torch.div(torch.mm(x, x.t()), torch.mm(norm2, norm2.t()) + 1e-7) RuntimeError: expected device cpu but got device cuda:0
all the code is your, but the train.py I used from the kif's GCN with little modification. But I want to ask you a question, Can you explain your model.py...
mldl@ub1604:~/ub16_prj/pytorch-AGNN/agnn$ python train.py Loading cora dataset... Traceback (most recent call last): File "train.py", line 113, in train() File "train.py", line 23, in train output = model(features, adj) File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line...