Problems encountered running the train.py file

[mh8859]

Hello，When I run the line loss.backward(), I get an error:

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [32, 1000, 256]], which is output 0 of ReluBackward0, is at version 1; expected version 0 instead. Hint: the backtrace further above shows the operation that failed to compute its gradient. The variable in question was changed in there or anywhere later. Good luck!

The first time I encounter this error, it seems that the error message is caused by the relu function.I really don't know how to solve this problem, due to server graphics card issues, I can not use Torch 1.6.0, must choose a higher cuda version (Torch 1.13.0), I don't know if this is the reason. I am looking forward to your reply and hope to get your help

[LirongWu]

This is a problem due to pytorch version, the reasons behind it are complex and you may find some solutions here https://www.google.com/search?q=inplace+operation+pytorch.

[mh8859]

Thank you very much for your answer. Originally, I thought that the source code of Relu incentive function had changed. According to your help, I successfully solved the problem.

In addition, the data sets you provide are all data content encoded by numpy, can you upload a copy of the original data content (or part of the data), so that we can more intuitively understand what data your model works on?

[LirongWu]

You can refer to https://github.com/Shen-Lab/CPAC to see what data is and how to process it.

[jinzhuwei]

Thank you very much for your answer. Originally, I thought that the source code of Relu incentive function had changed. According to your help, I successfully solved the problem.

In addition, the data sets you provide are all data content encoded by numpy, can you upload a copy of the original data content (or part of the data), so that we can more intuitively understand what data your model works on?

I would like to ask you how to modify this problem. I hope to get your help