Unanswered
Question About The Usage Of Trains Agents.
In Our Company We Have 3 Hpc Servers, Two Of Them Have Multiple Gpus, One Is Cpu Only.
I Saw In The Docs The Multiple Agents Can Be Run Separately Assigning Gpus In Whatever Manner You Want.
My Questions Are
1
So I assume, trains assumes I have nvidia-docker installed on the agent machine?
docker + nvidia-docker-runtime are assumed to be installed
nvidia/cuda docaker image is pulled when requested (like any other container image)
Moreover, since I'm going to use
Task.execute_remotely
(and not through the UI) is there any code way to specify the docker image to be used?
Sure, task.set_base_docker(docker_cmd='nvidia/cuda -v /mnt:/tmp')
Notice that you can not only pass the docker image but also provide the docker with execution parameters like volume mounts or environment variables, etc.
132 Views
0
Answers
4 years ago
one year ago
Tags