Badges 15 × Eureka!
Thanks for the answer!
the idea is to use ClearML for the node allocation/scheduling and PyTorch DDP for the actual communication, is that correct ?
Yes, i basically plan to use ClearML as user-friendly cluster manager
Regrading Task pollution, when the master node is done, it can delete all child/cloned Tasks so it is easier on the eyes.
Nice idea, i will try it out!
I am referring to server-side setup from here https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_security#example-using-environment-variables
Thanks! It means all of these are redundant if I use clearml.conf?
Hmm, not sure. I guess i lack web security knowledge. When are those keys used?
Aha! That makes sense. Thanks a lot guys!