I installed the following versions of dependencies as required,Python == 3.6, cuda == 10.2,Pytorch==1.5.1,PyG: torch-geometric==1.5.0,but However, the following error is reported:RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED,my guess is that maybe the version of the package is not correct.So I reconfigured the environment as follows:pytho ==3.9,pytorch==1.13.1,cuda==10.6,torch-geometric ==2.2.0,but then the following error occurs:
File "/home/isac/shared_workspace/chengyuanyuan_new/GDN_main/models/graph_layer.py", line 110, in message
alpha = softmax(alpha, edge_index_i, size_i)
File "/home/isac/anaconda3/envs/GNN_new/lib/python3.9/site-packages/torch_geometric/utils/softmax.py", line 56, in softmax
ptr = ptr.view(size)
AttributeError: 'int' object has no attribute 'view'
I installed the following versions of dependencies as required,Python == 3.6, cuda == 10.2,Pytorch==1.5.1,PyG: torch-geometric==1.5.0,but However, the following error is reported:RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED,my guess is that maybe the version of the package is not correct.So I reconfigured the environment as follows:pytho ==3.9,pytorch==1.13.1,cuda==10.6,torch-geometric ==2.2.0,but then the following error occurs: File "/home/isac/shared_workspace/chengyuanyuan_new/GDN_main/models/graph_layer.py", line 110, in message alpha = softmax(alpha, edge_index_i, size_i) File "/home/isac/anaconda3/envs/GNN_new/lib/python3.9/site-packages/torch_geometric/utils/softmax.py", line 56, in softmax ptr = ptr.view(size) AttributeError: 'int' object has no attribute 'view'