Inplace operation resulting in the code failing to run

[HanbingWang2001]

When the code runs to Phase II, an error occurs as follows.
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [100, 1]], which is output 0 of TBackward, is at version 2; expected version 1 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).
Thank you very much!

[CharlieHu99]

Hi, thanks for the question. This error may be caused by a different version of PyTorch, the code works properly on PyTorch version 1.4.0.

[HanbingWang2001]

Thanks!