Is it possible to clean the model when you are done auto annotating?

I'm launching label-studio and label-studio-ml-backend on the same Docker container and the only way I've found to clear the resources ml-backend is using is to stop the container. I just want to clear the ml-backend and keep the label-studio server running. Is there some way to do that?

Thanks in advance.

The issue names no files, tests, or entry points. Start by examining how label-studio-ml-backend and Label Studio run together in the same Docker container, then determine whether the backend can be cleared independently. Done means identifying and documenting a supported way to reset the ML backend while keeping the Label Studio server running.

Written by the indexing model from the issue text.