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
I can see on my platform monitoring that during training, only 1 vCPU out of 8 is used. Apart from setting num_workers parameters in params.jl (which I left at the default value of 128 for all occurrences), is there something to be done in order to effectively use multiple cores?
The text was updated successfully, but these errors were encountered:
Originally posted by @jonathan-laurent in #18 (comment)
I can see on my platform monitoring that during training, only 1 vCPU out of 8 is used. Apart from setting
num_workers
parameters inparams.jl
(which I left at the default value of128
for all occurrences), is there something to be done in order to effectively use multiple cores?The text was updated successfully, but these errors were encountered: