I understand for cleaml-agent
What I mean is that I have 2 self deployed server. I want to switch between the 2 config when running the code locally, not inside the agent
nice !! That is exactly what I am looking for !!
can you make train1.py
use clearml.conf.server1
and train2.py
use clearml.conf2
?? In which case I would be intersted @<1523701087100473344:profile|SuccessfulKoala55>
You can use the CLEARML_CONFIG_FILE
environment variable
That's not trivial, since the SDK uses a pre-initialized session to communicate with the server. Switching means changing this session and making sure all cached instances of the session are dropped. It would be much easier just to use two different scripts
Since an experiment is run by an already running agent, the configuration of the agent can only be controlled when you actually start the agent