Implementing on Jetson Xavier NX

[rmadan2]

I would like to begin by saying that this is indeed an exceptional effort. I am impressed by the robustness and stability of DROID-SLAM in particular. I am currently trying to implement DROID-SLAM on a Jetson Xavier NX. I understand that a GPU with at least 11GB of memory is needed for inference, but I am not currently interested in real-time performance. The Jetson Xavier NX board that I am using currently is running with Python 3.6, with CUDA 10.2 and Pytorch 1.10, and I am getting the following error:

[rmadan2]

I understand that an 801 error means that the service isn't currently supported on the NVIDIA device. but is there a work-around to it?

[surajiitd]

@rmadan2 did you resolved this issue?, I also got the same problem in Jetson Xavier AGX. Could you please help!

[surajiitd]

And when I passed --disable_vis flag, then the new error is like this: RuntimeError: CUDA error: no kernel image is available for execution on the device

My output for torch.cuda.is_available() is True

My output for torch.cuda.get_arch_list() is ['sm_53', 'sm_62', 'sm_72']

My output for torch.Tensor([10,23]).to("cuda") is tensor([10., 23.], device='cuda:0')

[rmadan2]

Hi @surajiitd , I have not been able to resolve this issue