As the code is expected to grow and be refactored in the next months, it would be important for us to develop abstract classes for data generation, fitting, benchmarking and auto workflows combining all 3.
Similarily, document models for such flows need to be developed to ensure easy connection of the auto workflow parts.
I have started to work on this and will post a draft within the next weeks.