I have a production inference pipeline which I want to continuously test on my Github to make sure it doesn't break as we move forward.
The ideal scenario for me is to use https://docs.github.com/en/actions/using-containerized-services/about-service-containers in my Github action flows to launch self-deployed ClearML server and test the pipeline there. But the problem is that setting up a ClearML server requires running some command line commands (like creating the directories etc.) which I cannot do in the services section.
How would you recommend continuously testing processes that are ClearML based?
Well, ideally you simply have a super-light server deployed somewhere - that's the easiest solution.
I don't think setting up a server each time is a good solution, however it might be possible to do so without the external configuration, assuming you don't mind the server's data not being persistent...
Gotcha, didn't think of an external server as Service Containers are part of Github's offering, I'll consider that
