Hi AttractiveCockroach17 , in the first question - clearml captures the packages used during the run. What does your script use and what does clearml capture when running locally on your machine?

You can configure the clearml to capture your entire environment as well.

Regarding 2:
Can you please expand on the entire process?

1.- The script im running uses qiskit.providers but as installed by when you install qiskit. If you try to install the submodules independently, it doesnt work. How do I use the full environment instead? cannot find this in the documentation. Also, I cannot configure the agents it seems because im using the aws autoscaler service so I dont spin them explicitly.

2.- My workflow would be that I usually, locally I run multiple sequential experiments using hydra multirun. What I want is that If I pass an extra flag, like, remote=true , all these runs get enqueued. So at first thought excecute_remotely would work but doenst do what I need because it stops the full excecution, not only one of the runs from hydra's multirun, but the whole thing.

found the env freeze. For the second workflow all I would need I guess then would be and env variable that would tell me whether this is being currently run by an agent or not

basically running_locally() ok, I think I have everything I need. Will give it a try.