RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED

[cyanling2]

Hello, I am trying to reimplementing the pattern branch and have this issue. There is also a warning saying that "GeForce RTX 3070 with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 compute_37." I am using Ubuntu 21.04 and CUDA of version 11.2. I this the error is due to incompatibility between my rtx3070 and pytorch 1.6. I think I need a combination of pytorch=1.8, torchvision=0.9 to get it work. Do you have an alternative environment for this? ty!

[PeterTF656]

same here +1, seems like that 30 series GPU is not compatible with the environment...

[thaoshibe]

Sorry for the inconvenience (⌣̩̩́_⌣̩̩̀) I trained and tested the code on RTX 2080Ti, CUDA 10.2, Ubuntu 18.04; I haven't tried any other alternative environment. So all help is much appreciated and very welcome!

[shiwen-cheng]

hello， I'm using RTX 2080Ti, CUDA 10.2, Ubuntu 16.04, cuDNN8.3.0; and I'm facing the samely issue: a call to cuDNN failed. Maybe it's not right version of cuDNN? Can I ask what version of cuDNN do you use and your NVIDIA driver version? I think this error may be related to these.