Frederik
Frederik
He meant the compute capability of the GPU. My GPU has compute capability 7.5, so I set `export TORCH_CUDA_ARCH_LIST="7.5+PTX"` and it worked
This leads to the same result. It tries installing with CUDA 12.x and PyTorch 2.1.x
I am having the same problem. I also replaced the Franka Panda with a different robot but even if I put the Panda back in it still does not work....
It seems to be resolved for now. Could be due to having restarted the docker container or could be due to the CUDA kernel fix - I will see. Either...
The problem came back. Could you please reopen the issue? I first ran my code without problems. Then I wanted to check the result of an earlier computation in the...
My dockerfile is based on the one provided in this repo, so the command is `pip install .[dev] --no-build-isolation`. Also, I played around with the IK solver and noticed that...
Same, got the exclamation mark on the icon but everything else works as expected.