Stratification Bug while using apex(pullback(...))

[Tom-Szendrey]

While trying to use apex(pullback(...)) or ob(pullback(...)) sname is the only attribute populated Below is what was ran to reproduce this result. Examples came from both demo notebooks

I was able to get the expected result using:

res = stratify(modelATyped=>[[:strata],[:strata],[:strata],[]], # S I R D
        modelBTyped=>[[:disease], [:disease,:infect]],# Q NQ 
        typesP
) 

function stratify(petri_net1, petri_net2, type_system)
  return pullback([add_cross_terms(petri_net, type_system) for petri_net in [petri_net1, petri_net2]]) |> apex
end

[kris-brown]

if you're following the stratification made easy demo, then you are getting expected results.

The pullback you are trying to construct was called naive_stratification in the notebook, and it was an example intended to demonstrate that "directly" stratifying those two models will produce a model that has no transitions. (nt counts the number of transitions and the notebook showsnt(naive_stratification) == 0). The stratify helper function automates how you have to update the SIRD and Quarantine models such that the stratification that we really want is computed via Catlab's pullback.

Let me know if any of the wording in that notebook can be fixed or if this doesn't actually address your problem!

[Tom-Szendrey]

Thanks! This clears things up. I believe I was misunderstanding due to some online chats we had I will continue using the stratify helper function moving forward