CUDA error: no kernel image is available for execution on the device Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

OpenGVLab / VideoMamba

VideoMamba: State Space Model for Efficient Video Understanding

https://arxiv.org/abs/2403.06977

Apache License 2.0

660 stars 47 forks source link

CUDA error: no kernel image is available for execution on the device Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. #41

Closed ZXXaaaa closed 2 months ago

ZXXaaaa commented 2 months ago

CUDA error: no kernel image is available for execution on the device Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

Have you ever encountered this problem? Do you know how to solve it?

Andy1621 commented 2 months ago

Hi! Could you give more details about your GPU and environment?

ZXXaaaa commented 2 months ago

The previous GPU was TITAN XP, and pytorch2.1.1 torchvision0.16.1 torchaudio2.1.1 cuda11.8 python3.10.13. It may be the problem of computing power. I rented the cloud server, and now the problem has been solved. Thank you for your attention and excellent work!