does this code support linux GPU training host?

[qingyuan18]

The repo is using windows cuda dll lib , which can't run on linux gpu marchine

Is there any roadmap to support linux training hosts? otherwise we need to adapt the code ourselves

Thanks

[delight16]

The repo is using windows cuda dll lib , which can't run on linux gpu marchine

Is there any roadmap to support linux training hosts? otherwise we need to adapt the code ourselves

Thanks

Currently, I have also encountered this problem. May I ask how you resolved it?

[XiangguanMu]

I comment the

import os
os.add_dll_directory('c:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.1/bin')
os.add_dll_directory(os.path.dirname(__file__)) 

in model_rd.py, and install

torch                 1.9.0+cu111
torch_geometric       2.0.2
torch-scatter         2.0.7
torch-sparse          0.6.10

, then it works for my linux server.