[muon id] bug fix for high pt muon algorithm ("Picky")

[drkovalskyi]

From @rradogna:

the fix is for the piece of logic that tries to figure out whether the muon fit is using segments or hits - because picky counts hits in individual layers when it looks for showers, so if the fit is using segments it has to get their component rechits to look at individual layers. For some reason it was checking the size() of the rechit to see if it is a hit (size ==1) or segment (size>1). And it failed for 4d segments that have only one phi/theta projection - then size()==1 and the code thinks it is a segment. So I changed this to checking the hit dimension which makes much more sense.

FYI @ptraczyk

[cmsbuild]

The code-checks are being triggered in jenkins.

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-22422/3675

[cmsbuild]

A new Pull Request was created by @drkovalskyi for master.

It involves the following packages:

RecoMuon/GlobalTrackingTools

@perrotta, @cmsbuild, @slava77 can you please review it and eventually sign? Thanks.
@bellan, @abbiendi, @jhgoh, @echapon, @calderona, @HuguesBrun, @folguera, @battibass, @trocino, @amagitte, @bachtis, @rociovilar this is something you requested to watch as well.
@davidlange6, @slava77, @fabiocos you are the release manager for this.

cms-bot commands are listed here

[slava77]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/26450/console Started: 2018/03/02 17:05

[cmsbuild]

+1
Tested at: 4eb5336
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22422/26450/summary.html

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22422/26450/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 0 differences found in the comparisons
• DQMHistoTests: Total files compared: 29
• DQMHistoTests: Total histograms compared: 2479021
• DQMHistoTests: Total failures: 1
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2478844
• DQMHistoTests: Total skipped: 176
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 1.23999999996 KiB( 23 files compared)
• Checked 118 log files, 9 edm output root files, 29 DQM output files

[perrotta]

@drkovalskyi @rradogna
There are no changes in the jenkins outputs: shouldn't we expect some?
Did you study the effect of the fix? Maye a larger stat than the automatic tests is needed to appreciate it?

[drkovalskyi]

I would not expect it to be visible easily. The algorithm is relevant only at high pt. Raffaella is not available at the moment. Maybe Piotr can comment (@ptraczyk)

[perrotta]

Tested with 800 events SingleMuPT1000 with 2018 config (wf 10809, 1600 muons at 1 TeV pt).

Two muons get their pt modified with the picky algo (and the new pt is closer to 1 TeV than before, whatever it means with such a low stat):

The number of picky muons whose chi2 has been modified (typically reduced) is a bit larger: 5 muons (out of 1600):

Touching the chi2 (and recHit multiplicities, etc.) leads to some other reco::muon (besides the two whose picky muon pt got modified) whose momentum assignment gets modified:

(in this plot the pt is cut at 1 TeV, which is rather unfortunate here: this doesn't want and can't be a complete validation of the fix, though, and the plot as a whole just testify that there are more reco::muon affected than than the two muons which got a different momentum measured by the picky algo)

[perrotta]

+1

• The changes in the code are in line with what declared
• The effect of the fix is limited to a small fraction of the high pt muons: it can propagate to other derived quantities (e,g, jets, MET, etc.) in case the muon momentum assignment is also affected
• At a first glance, those changes go in the correct direction, although some larger stat comparison than the limited one inspected in [muon id] bug fix for high pt muon algorithm ("Picky") #22422 (comment) is needed for a complete validation

[cmsbuild]

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. @davidlange6, @slava77, @smuzaffar, @fabiocos (and backports should be raised in the release meeting by the corresponding L2)

[fabiocos]

+1