dectrfov/Wavelet-U-net-Dehazing

a problem with the backpropagation process

Opened this issue · 5 comments

cloudyliy commented

The following error occurred when running the code：

File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/tensor.py", line 102, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/autograd/init.py", line 90, in backward
allow_unreachable=True) # allow_unreachable flag
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

It seems that there is a problem with the backpropagation process. Has the author encountered such a problem? My pytorch version is 1.1.0, thank you!

dectrfov commented

Thanks for your interest.
Can you show the detailed error message?
In our machine, I don't encounter this condition.

Best Regards

luoxiaodou commented

The following error occurred when running the code：

File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/tensor.py", line 102, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/autograd/init.py", line 90, in backward
allow_unreachable=True) # allow_unreachable flag
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

It seems that there is a problem with the backpropagation process. Has the author encountered such a problem? My pytorch version is 1.1.0, thank you!

I also encounter the problem, do you solve it? Could you tell me the solutions?

zj1010 commented

The following error occurred when running the code：
File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/tensor.py", line 102, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/autograd/init.py", line 90, in backward
allow_unreachable=True) # allow_unreachable flag
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation
It seems that there is a problem with the backpropagation process. Has the author encountered such a problem? My pytorch version is 1.1.0, thank you!

I also encounter the problem, do you solve it? Could you tell me the solutions?

I met the same problem，do you solve it?

Ypigeon commented

运行代码时出现以下错误：文件“/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/tensor.py”第102行，向后torch.autograd.backward（自我，渐变，保留_graph，create_graph）文件“/home/xxx/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/autograd/init.py”，第90行，向后allow_unreachable=True) # allow_unreachable标志RuntimeError：梯度计算所需的一个变量就就地修改反向传播过程似乎存在问题。作者遇到过这样的问题吗？我的pytorch版本是1.1.0，谢谢！> >> >> > 我也遇到了这个问题，你解决了吗？你能告诉我解决方案吗？>> 我遇到了同样的问题，你解决了吗？

xiaowowwwww commented

Hello, I also have the above problem, I would like to ask if there is any solution? Thank you！
Traceback (most recent call last):
File "D:\Wavelet-U-net-Dehazing-master\train.py", line 130, in
main(config_args)
File "D:\Wavelet-U-net-Dehazing-master\train.py", line 96, in main
loss.backward()
File "D:\anaconda\envs\pytorch\lib\site-packages\torch_tensor.py", line 488, in backward
torch.autograd.backward(
File "D:\anaconda\envs\pytorch\lib\site-packages\torch\autograd_init_.py", line 197, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [16, 4, 64, 64]], which is output 0 of AsStridedBackward0, is at version 16; expected version 15 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!