search

TrustAGI-Lab / ARGA

This is a TensorFlow implementation of the Adversarially Regularized Graph Autoencoder(ARGA) model as described in our paper: Pan, S., Hu, R., Long, G., Jiang, J., Yao, L., & Zhang, C. (2018). Adversarially Regularized Graph Autoencoder for Graph Embedding, [https://www.ijcai.org/proceedings/2018/0362.pdf].
MIT License
185 stars 58 forks source link

About the use of GPU #13

Open DreamerDW opened 4 years ago

DreamerDW commented 4 years ago

Hi Ruiqi, Thanks for your nice work! Does your code support running on GPU? I haven't modified it successfully.

Thank you very much!

yliuhz commented 2 years ago

I solved the problem by simultaneously modifying the "CUDA_VISIBLE_DEVICES" setting in two .py files, namely "clustering.py" and "link_prediction.py", since the main "run.py" file import both of them. I guess that you only modified either one of them.