Skip to content

write configuration file from bamboo for use in unfolding scripts

Kyle Cormier requested to merge kcormier/ttbbRun2Bamboo:config_writing into master

The idea here was that when we define the histograms, we also write out some information about them that can be used as configuration in the unfolding setup afterwards. Writing out two configuratoin files, one the unfolding_config which relates the gen-level plots to the migration/reco levels and contains some extra information. The other unfolding_plots which contains information about each of the reco-level histograms, including the auxiliary observables used (also stored in this file). Currently this doesn't include any information about e.g. systematics, though we can consider if those should be added in. Example config files from a short test run: unfolding_plots.yml


This is a work in progress and general feedback on how useful (or not) it is, what we might want to include is welcome.

Conceptually this also re-arranges the code so that first one defines the collections and selections that will be used, and then defining all plots (migration, reco and gen) at the same time. I was worried this would be quite slow based on previous comments, but some short testing with a data and signal file didn't show any reduction in speed compared to what was running previously.

Edited by Kyle Cormier

Merge request reports