Hi Everyone, I Am Trying To Use Docker Mode For Trains-Agent, But It Seems That It Has Problem With The Use Of Multiple Gpus This Is My Trains-Agent Command: Trains-Agent Daemon --Gpus 0,1 --Queue Dual_Gpu --Docker --Foreground And It Gets The Error: Doc

Unanswered

when I launch this:
(trains-agent) lv-beast@lv-beast:~/dev/MachineLearning/scripts/cmd_launcer$ docker run --gpus '"device=0,1"' nvidia/cuda:10.1-base nvidia-smi
it worked, so maybe its an issue with how trains pass the device to the docker run command?

  				
Posted 
	3 years ago

					More
				  		
  Report
		
					RattySeagull0
				
					0
					 × 1

84 Views

0 Answers

3 years ago

one year ago