RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

[YuheD]

/DMIT/models/base_model.py", line 246, in update_model errDec_total.backward(retain_graph=True) RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [8, 256]] 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).

Anybody has this question?

[lknownothing]

I encountered a same problem as yous. My pyotrch version is 1.7.1+cu101. Have you solved the problem

[Xiaoming-Yu]

Please downgrade your pytorch version to pytorch<=1.4. I will refactor the code in the near future to support the new version of pytorch.

[Idea89560041]

你好，作者，我也遇到相同的问题，目前使用的是pytorch1.7，当前的显卡最低就支持CUDA11，支持的最低版本Pytorch就是1.7，很难降低版本，请问作者有什么好的解决方案吗，谢谢！