Unable to backpropagate due to inplace operations

FelixOpolka / STGCN-PyTorch

🚗 Implementation of spatio-temporal graph convolutional network with PyTorch

MIT License

335 stars 70 forks source link

Unable to backpropagate due to inplace operations #4

Closed jeffreyjeffreywang closed 5 years ago

jeffreyjeffreywang commented 5 years ago

When running main.py, a runtime error occurs: one of the variables needed for gradient computation has been modified by an inplace operation. What are the possible lines that cause this issue? Is it line 83 (torch.einsum) in stgcn.py?