Could you please tell me how to fix the bug?

[qiaoranliqu]

File "UAI1_full_resolution.py", line 278, in
out = model(batch)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "UAI1_full_resolution.py", line 36, in forward
x = F.relu(self.conv1(x, edge_index, edge_attr))
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/graph-pde-master/graph-neural-operator/nn_conv.py", line 271, in forward
return self.propagate(edge_index, x=x, pseudo=pseudo)
File "/root/miniconda3/lib/python3.8/site-packages/torch_geometric/nn/conv/message_passing.py", line 317, in propagate
out = self.message(**msg_kwargs)
File "/root/graph-pde-master/graph-neural-operator/nn_conv.py", line 274, in message
weight = self.nn(pseudo).view(-1, self.in_channels, self.out_channels)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/graph-pde-master/graph-neural-operator/utilities.py", line 226, in forward
x = l(x)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/activation.py", line 98, in forward
return F.relu(input, inplace=self.inplace)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/functional.py", line 1442, in relu
result = torch.relu(input)
RuntimeError: CUDA out of memory. Tried to allocate 1.44 GiB (GPU 0; 23.70 GiB total capacity; 20.41 GiB already allocated; 18.56 MiB free; 21.69 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

We implement the code on Cloud Platform and the cost of GPU exceeds 24GiB, so could you tell us the capacity of your GPU?

[dericknganyu]

Reduce the batch size you use per epoch. This should solve the issue

[freebob]

I also encountered the same problem, how did you solve it?thx

[bazziamir24]

Reduce the batch size and try to clean up catch in the CUDA manually.