train.py problem??

[DavidKong96]

sorry,when i run train.py, there is a bug
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [8, 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).
could you please tell me why that happend?

[hytseng0509]

Thanks for raising the problem. Which Pytorch version are you using?

[DavidKong96]

Thanks for raising the problem. Which Pytorch version are you using?

i use 1.5.0

[hytseng0509]

Can you try to run the process using Pytorch 1.4?
We are correcting the problem for Pytorch 1.5.

[DavidKong96]

Can you try to run the process using Pytorch 1.4?
We are correcting the problem for Pytorch 1.5.

yes,it works on 1.4.0 ! thank you so much

[SmartNight-cc]

Can you try to run the process using Pytorch 1.4? We are correcting the problem for Pytorch 1.5.

Hello! I have the same problem, my torch.version=1.11.0, how can I fix it?

[fikry102]

Can you try to run the process using Pytorch 1.4? We are correcting the problem for Pytorch 1.5.

Hello! I have the same problem, my torch.version=1.11.0, how can I fix it?

I also had the same problem. My torch.version is 1.8.0.
I find this problem may have something to with the torch.autograd.grad issue #40.
I create a QQ group whose group number is 693337454. Anyone interested in these problems is welcome to join in us for further dicussion.

[SmartNight-cc]

Can you try to run the process using Pytorch 1.4? We are correcting the problem for Pytorch 1.5.

Hello! I have the same problem, my torch.version=1.11.0, how can I fix it?

I also had the same problem. My torch.version is 1.8.0. I find this problem may have something to with the torch.autograd.grad issue #40. I create a QQ group whose group number is 693337454. Anyone interested in these problems is welcome to join in us for further dicussion.

Hello, I did not search for this group number

[fikry102]

Can you try to run the process using Pytorch 1.4? We are correcting the problem for Pytorch 1.5.

Hello! I have the same problem, my torch.version=1.11.0, how can I fix it?

I also had the same problem. My torch.version is 1.8.0. I find this problem may have something to with the torch.autograd.grad issue #40. I create a QQ group whose group number is 693337454. Anyone interested in these problems is welcome to join in us for further dicussion.

Hello, I did not search for this group number

I just forgot to enable searching the group by group number. You can try it again.

[juliusyang97]

Can you try to run the process using Pytorch 1.4?
We are correcting the problem for Pytorch 1.5.

yes,it works on 1.4.0 ! thank you so much

pytorch=2.0.0 is works, you can try