TAO toolkit docker: Error response from daemon: could not select device driver "" with capabilities: [[GPU]]

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Please provide the following information when requesting support.

• Hardware (GeForce RTX3080)
• Network Type (Classification ResNet 50)
• TLT Version TAO toolkit version 3.22.05)
• no Training spec file
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)

Hi I notice this error has come up quite a lot. The earlier posts all seem to have been closed.

I have disabled nouveau and secure boot. I have installed driver 510.73.05 and I’m running CUDA 11.6 (please see attached nvidia-smi for more), Ubuntu 18.04 on a dual boot ASUS ROG laptop.

I have also attached the nvidia-bug-report.log.gz
nvidia-bug-report.log.gz (342.6 KB)

nvidia-smi
nvidia-smi907×595 72.2 KB

Any advice will be very gratefully received
Thank you

2

Did you install nvidia-docker?

3

I did, but reinstalling nvidia-docker2 fixed things.

Many thanks

4

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.