-
Notifications
You must be signed in to change notification settings - Fork 9.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
RuntimeError: nms_impl: implementation for device cuda:0 not found. #6765
Comments
|
How to output mmcv environment like yours?
|
Run |
@EkAugust |
@daydayup-bit Then, I re-compiled mmcv directly. But I didn't know how to uninstall current mmcv building from source code. Thus, the recompilation failed. I can not find out the error reason. I delete the docker images and containers. Finally, I re-create my docker to solve it just like some unknown problems happen on your PC you and finally you reboot. |
hello, did you solve this problem? I also meet this problem, and I am sure I installed mmcv-full, but it is also not work.
TorchVision: 0.11.3 |
you guys all got 'MMCV CUDA Compiler: not available' |
I got this error too. |
install mmcv-full use |
In docker containers, we need to set environment variables as follows before executing pip install:
This solved the problem for me |
Thanks for your help! By using those flags when installing the MMCV project from source also worked for me! However, when doing it in a Dockerfile, the CUDA runtime was not recognized and the docker build fails. Which hardware are you using, @jeethesh-pai ? I'm using a Jetson AGX Xavier. |
I have the same problem when using Setting these ENV variables in Dockerfile doesn't help: ENV FORCE_CUDA="1"
ENV MMCV_WITH_OPS=1 Example: from mmcv.ops import batched_nms
import torch
def check_mmcv():
device = torch.device('cuda:0')
bboxes = torch.randn(2, 4, device=device)
scores = torch.randn(2, device=device)
labels = torch.zeros(2, dtype=torch.long, device=device)
det_bboxes, keep_idxs = batched_nms(bboxes.to(torch.float32), scores.to(torch.float32), labels, {
'type': 'nms',
'iou_threshold': 0.6
})
print('OK.')
if __name__ == '__main__':
check_mmcv() Error:
|
Same results for me. I am installing
|
any updates on that? I'm getting the same issue with running @rayryeng snippet with env:
|
Thank you jshilong
If you get the following error:
Then uninstall both |
@aRibra MMCV v2.0.0 official version was released on April 6, 2023. In version 2.x, it removed components related to the training process and added a data transformation module. Also, starting from 2.x, it renamed the package names mmcv to mmcv-lite and mmcv-full to mmcv. For details, see Compatibility Documentation. |
I installed mmcv with |
Hi,
I don't know how to solve the issue.. |
it works, thank you |
I was getting the same error on mmcv 2.1.0 and this fixed it for me. <tldr;> uninstall mmcv then reinstall with the command Details:
Missing from
When I did this, I was getting:
Adding didn't seem to have an impact... BUT What actually fixed it was: |
When I run
/usr/src/app/demo/inference_demo.ipynb
, an error reported:My environment:
The text was updated successfully, but these errors were encountered: