RuntimeError

[Narges1996]

Hi
When I want to run 'train_Stanford_Cars_TResNet_L.py', I get this error:
RuntimeError: Some elements marked as dirty during the forward method were not returned as output. The inputs that are modified inplace must all be outputs of the Function.
How can I solve that?
my torch version is 2.0.1+cu118
my torchvision version is 0.15.2+cu118

[Dichao-Liu]

@Narges1996
Hi, thank you very much for your interest in my work.

I tried to use torch 2.0.1, and I also had the same problem. I suppose it might be some version conflict. You may try older pytorch, or ask the authors of Tresnet for help.

[nobody2001]

Hi
When I want to run 'train_Food_11_ResNet50.py', I get this error:
"RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions."

Have you ever run into this problem?
How can I solve that?
my torch version is 2.0.0+cu118
my torchvision version is 0.15.1+cu118