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

[cc421528]

Help: Have any classmates encountered this problem? I don't know how to solve it.
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [27, 32]], 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!

[wang-jing1]

帮助：有没有同学遇到过这个问题？我不知道如何解决它。运行时错误：梯度计算所需的变量之一已被就地操作修改：[torch.cuda.FloatTensor [27， 32]]，这是 ReluBackward0 的输出 0，位于版本 1;预期版本 0 改为。提示：上面的回溯显示了计算梯度失败的操作。有问题的变量后来在那里或任何地方都发生了变化。祝你好运！

Yes, I'm having the same issue and don't know how to fix it yet.

[cc421528]

Hello, I have successfully resolved my issue: when I encountered this issue, my Pytorch version was 1.4. When I changed its version to 1.5, the issue was successfully resolved. I hope it can inspire you.

[wang-jing1]

您好，我已经成功解决了我的问题：当我遇到此问题时，我的 Pytorch 版本是 1.4。当我将其版本更改为 1.5 时，问题已成功解决。我希望它能激励你。

Thank you very much for your help. I can't wait to follow your instructions. Best wishes.