This repository has been archived by the owner on Sep 18, 2024. It is now read-only.
Replies: 1 comment
-
Actually, I found a trick. By duplicating the remote multiple time in the machinelist and set maxTrialNumberPerGpu:1 for each one. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
There are 4 remote ssh machines and each one has one GPU.
I try to set
trialGpuNumber: 1
trialConcurrency: 16
And then for each machine in machineList:
I have:
maxTrialNumberPerGpu:4
useActiveGpu: true
I expected there would be 16 trails running parallel, 4 on each.
But actually, 4 are running and 12 are waiting.
Beta Was this translation helpful? Give feedback.
All reactions