%def: "displayCompression" "none" "weak" « strong"
Basicaly: « none » will prompt any event that are causality related to the tracked event, without any compression
« weak » will extract a minimal scenario
« strong » will consider agents up to permutation, which induces a lot of computational time, but may reduce the scenario further.
Examples may be found in the git repository: tests/integration/cflows
You may also have a look at the resources of my M2 course (including the slides and some toy examples).
See Session 4 in https://www.di.ens.fr/~feret/teaching/2024-2025/MPRI.2.19/
Would you need further help, please let us know, or if you have specific development needs.
Best.
Jérôme.