Skip to content

Centralize the definition of make_jets to standard_jets.py

Dillon Scott Fitzgerald requested to merge df-centralize-make_jets into qee_upgrade_24

The goal of this MR is to centralize the definition of make_jets to standard_jets.py

  • Removal of definition of make_jets from lines/qee/jets.py and lines/qee/qee_builders.py
  • Update imports accordingly
  • Pass in an object to make_jets for the jet tagging rather than a string.

I have tested over 5000 events using snippets/3017 and snippets/2995 and confirmed that I see the same exact numbers on my branch and the master branch for BW and rates for the full, turbo, and qee sprucing streams.

  • Full:
    • BW = 0.7752 GB/s
    • rate = 6.384 kHz
  • Turbo:
    • BW = 0.3648 GB/s
    • rate = 78.204 kHz
  • QEE's excl spruce:
    • BW = 0.251 GB/s
    • rate = 2.736 kHz

A similar exercise was done Thursday night (CEST) with the most recent qee_upgrade_24. Still had the exact same conclusion see: !3242 (comment 7816944) 👍

NB This can be merged after all other jet MRs for QEE to avoid any merge conflicts that may arise in the make_jets function. If any changes are made to the previous definitions of make_jets, I can update the definition in standard_jets accordingly in this MR.

Follow up issues: https://gitlab.cern.ch/LHCb-QEE/qee-hlt-run3/-/issues/34 + #744 (closed)

TO BE ASSESSED AND DEALT WITH AT A LATER DATE! KEPT FOR BOOKKEEPING USEFUL FOR THE JET TEAM

HOWEVER: I noticed some strange things when looking through the implementation of make_jets in qee_builders.py and hence lines/qee/vjets.py and lines/qee/top_muon_elec.py. make_jets in qee_builders uses make_topo_2body() as the tag object that is labeled 'SV', rather than make_topo_2body_with_svtag(). The latter is what we use in lines/qee/jets.py, so for now there is a bit of a mismatch (or at least mislabeling which is confusing). I worry this may be an oversight... Do we want to fix this?? If so, I could put this fix in another MR as it could have some effect on the BW and rates (we can discuss this in our jet meeting on 03 April 2024). For now, I have just changed the labeling from SV to TOPO in lines/qee/vjets.py and lines/qee/top_muon_elec.py --

Edited by Luke Grazette

Merge request reports