-
Notifications
You must be signed in to change notification settings - Fork 21.4k
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
Build the docker image from source, but torch.cuda.is_available()==false #12773
Comments
@linkerr could you also run the |
@weiyangfb Hi, first of all, thank you, the collect_env.py run result: |
Please test compiling a CUDA sample and running it in the docker image. |
I'm having the same problem.
Running CUDA sample on image:
Related to #21259? |
Are you building images with kaniko or dockerindocker? |
We face the same issue. The pytorch installation from source is done similar to: https://github.com/pytorch/pytorch/blob/v1.5.0/docker/pytorch/Dockerfile
The container was build on a system without CUDA. Not sure if this is an issue. There were no special flags set for the build process. Not sure if that is necessary. From within the container |
@Spenhouet Have you specified you compiling options(sm50 like stuff)? My cuda extension compiles after setup up those things. |
@Godricly There were no further configurations than the above Dockerfile. Where would I set these compiler options? |
@Spenhouet If you are compiling C++ extension, you can setup them with CUDA compiler flags. Like the setup.py file. |
@Godricly I provided the dockerfile above. Could you please provide a dockerfile including these compiler flags that you are mentioning? |
❓ Questions and Help
Please note that this issue tracker is not a help form and this issue will be closed.
We have a set of listed resources available on the website. Our primary means of support is our discussion forum:
System info Tesla P100
cuda test show pass
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 9.0, CUDA Runtime Version = 9.0, NumDevs = 8
Result = PASS
but pytorch can't find cuda device:
I have no idea, please help, thank you!
The text was updated successfully, but these errors were encountered: