I'm running some ML python code on Amazon's EC2 - the machine has a GPU: NVIDIA A10G. The ML code is written using PyTorch.

When I run the code I get error:

CUDA error: no kernel image is available for execution on the device

I suppose this is about PyTorch version? Used version is torch==1.10.2.

How do I know which version of PyTorch the code MUST use? What's the error about?

Are there certain requirement for CUDA by this GPU (A10G)?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.