Hi again,
With regard to the previous posts in this thread, I still have one more question.
My motivation to use the overlapping model with discrete layers (instead of fitting a separate model to each layer) was my understanding that this model will somehow account for dependencies between the layers, and return partitions that (potentially) vary across layers. After reading through https://journals.aps.org/prx/abstract/10.1103/PhysRevX.5.011033, I'm still not sure whether/how my model accounts for such dependencies.
What is the difference between A) fitting a separate model to each layer and B) fitting an overlapping model with layers=True? When interpreting my results, what can I say about the dependencies between layers with discrete types of relationships?
Many thanks, Arttu