You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
/opt/nvidia/nvidia_entrypoint.sh: line 67: exec: .: cannot execute: Is a directory
Looks similar to #507 but the solution doesn't work.
I realised that I built the image before I installed the CUDA drivers and toolkit. Is there a need (and a way) to completely restart the process or is there another problem here?
The text was updated successfully, but these errors were encountered:
I had the same problem and I built the image after installing the CUDA drivers and toolkit. Turns out I had a trailing full stop on the line starting 'docker run'.
after building the container I get
Looks similar to #507 but the solution doesn't work.
I realised that I built the image before I installed the CUDA drivers and toolkit. Is there a need (and a way) to completely restart the process or is there another problem here?
The text was updated successfully, but these errors were encountered: