Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

tighter selection in pp_on_AA era for iterations still using pp defaults #39085

Merged

Conversation

mandrenguyen
Copy link
Contributor

In the heavy-ion era (pp_on_AA) each tracking iteration either has a dedicated MVA track selection, trained in heavy ions, or we simply tighten the cut on the pp-trained MVA output.
The exceptions are the jetCore and muon-seeded iterations, which are still using the pp defaults. The quality selection is far too loose for central PbPb collisions giving rise to fake tracks that can lead to fake hadrons in particle flow.
In the 2018 data this was dealt with by a post-PF cleaning module [1], which is sub-optimal, as the event interpretation is already done.
This PR tighten the selections that define the highPurity bit, which is required for charged hadrons in PF (at least for PbPb)

For the jetCore iteration we simply tightened the value of the pp-trained MVA cut used in the highPurity bit from 0.4 to 0.8. This was tested in minbias and QCD events (with a PbPb UE). There is about a 10% loss of tracks for this iteration, but the fake rate is reduced by a factor of several. The overall effect on the efficiency and fake rate is small, as this iteration does not contribute much to the overall tracking.

The muon-seeded iterations are still cut based. We tuned the cuts that define highPurity using a muon-jet and upsilon sample (with PbPb underlying event). There is no effect on the efficiency of muons passing the standard selection. This is expected as highPurity is not used in the global muon reconstruction or in the selection. Matched tracks from these iterations were reduced by about 5%, while fake tracks were reduced by 30%, notably removing a tail of high pT fake/mis-reco'd tracks, which was known to be problematic for jets.

@denerslemos

@cmsbuild
Copy link
Contributor

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-39085/31594

  • This PR adds an extra 16KB to repository

  • There are other open Pull requests which might conflict with changes you have proposed:

@cmsbuild
Copy link
Contributor

A new Pull Request was created by @mandrenguyen (Matthew Nguyen) for master.

It involves the following packages:

  • RecoTracker/IterativeTracking (reconstruction)

@jpata, @cmsbuild, @mandrenguyen, @clacaputo can you please review it and eventually sign? Thanks.
@VourMa, @felicepantaleo, @GiacomoSguazzoni, @JanFSchulte, @rovere, @VinInn, @ebrondol, @gpetruc, @mmusich, @mtosi, @dgulhan this is something you requested to watch as well.
@perrotta, @dpiparo, @qliphy, @rappoccio you are the release manager for this.

cms-bot commands are listed here

@jpata
Copy link
Contributor

jpata commented Aug 17, 2022

@cmsbuild please test

@cmsbuild
Copy link
Contributor

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-9f289f/26866/summary.html
COMMIT: 04db5e3
CMSSW: CMSSW_12_5_X_2022-08-16-2300/el8_amd64_gcc10
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/39085/26866/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 534 differences found in the comparisons
  • DQMHistoTests: Total files compared: 51
  • DQMHistoTests: Total histograms compared: 3692500
  • DQMHistoTests: Total failures: 17
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 3692461
  • DQMHistoTests: Total skipped: 22
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 50 files compared)
  • Checked 212 log files, 49 edm output root files, 51 DQM output files
  • TriggerResults: no differences found

@jpata
Copy link
Contributor

jpata commented Aug 17, 2022

+reconstruction

  • tightening HI cuts, with reco changes in HI
  • no reco impacts in pp

@cmsbuild
Copy link
Contributor

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @qliphy, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)

@perrotta
Copy link
Contributor

+1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants