check both algo and originalAlgo in electron track isolation calculators

[slava77]

similar to #35892, I'm proposing to check both the algo and the originalAlgo in track selections for EGM

@cms-sw/egamma-pog-l2
this was prompted by the observations in the EXO validation https://indico.cern.ch/event/1102815/
https://indico.cern.ch/event/1102815/contributions/4639510/attachments/2357228/4025487/W%27toenu%20mkFit%20vaildation%20study%20Run3.pdf
(page 5)

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-36364/27126

• This PR adds an extra 16KB to repository

[cmsbuild]

A new Pull Request was created by @slava77 (Slava Krutelyov) for master.

It involves the following packages:

• RecoEgamma/EgammaIsolationAlgos (reconstruction)

@jpata, @cmsbuild, @slava77 can you please review it and eventually sign? Thanks.
@Sam-Harper, @jainshilpi, @lgray, @sobhatta, @afiqaize, @wrtabb, @varuns23, @ram1123 this is something you requested to watch as well.
@perrotta, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

[slava77]

@cmsbuild please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-6f4b09/20988/summary.html
COMMIT: c0aec63
CMSSW: CMSSW_12_2_X_2021-12-04-1100/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/36364/20988/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: 60 differences found in the comparisons
• DQMHistoTests: Total files compared: 41
• DQMHistoTests: Total histograms compared: 3041955
• DQMHistoTests: Total failures: 132
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3041801
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 40 files compared)
• Checked 175 log files, 37 edm output root files, 41 DQM output files
• TriggerResults: no differences found

[slava77]

Reco comparison results: 60 differences found in the comparisons

these are, as expected from the code changes, coming from small increases in the track isolation

e.g in DoubleEG 2017C wf 136.793

or e.g. in miniAOD

@swagata87
please clarify if this PR needs additional checks.
Thank you.

[swagata87]

@swagata87
please clarify if this PR needs additional checks.
Thank you.

Hello Slava,
Track isolation is one of the input variables of the DNN for PF Electron ID (link to relevant code).
So, just to be fully sure, we are trying to check DNN performance in [Base] vs [Base+this PR]. In particular, we are trying to check if there is any impact on:

• Taus
• Particle flow in general

these checks are normally quite quick, and we expect to report back to you very soon (let's say 1-2 days).

[slava77]

@cmsbuild please test

these checks are normally quite quick, and we expect to report back to you very soon (let's say 1-2 days).

Thank you.
This timeline should work.

Since I made this PR to get more consistent performance with/without mkFIt, you may want to consider the no-mkFIt variant for comparison as well.

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-6f4b09/21125/summary.html
COMMIT: c0aec63
CMSSW: CMSSW_12_3_X_2021-12-09-1100/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/36364/21125/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: 58 differences found in the comparisons
• DQMHistoTests: Total files compared: 42
• DQMHistoTests: Total histograms compared: 3250704
• DQMHistoTests: Total failures: 126
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3250556
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 41 files compared)
• Checked 177 log files, 37 edm output root files, 42 DQM output files
• TriggerResults: no differences found

[swagata87]

We managed to complete the checks for PF Egamma ID DNN performance (Base vs Base+thisPR).

• Tau related checks are here , done by Mario Sessini from TAU POG. No degradation in performance after this PR.
• PF related checks are here, done by Ankush Reddy Kanuganti (@akanugan). There are many plots here. So, may be worth pointing out a few of them.

1. slimmedMET in ZEEPU: https://akanugan.web.cern.ch/akanugan/PF_Validation_PFEGMID_v2_DNNon_vs_DNNon+PR36364/ZEEPU_JetMETMETValidation/slimmedMETs/MET.pdf
2. slimmedMET in FlatQCD_PU: https://akanugan.web.cern.ch/akanugan/PF_Validation_PFEGMID_v2_DNNon_vs_DNNon+PR36364/FlatQCD_PU25ns_JetMETMETValidation/slimmedMETs/MET.pdf
3. chargedHadronEnergyFraction of slimmedJets in FlatQCD_PU: https://akanugan.web.cern.ch/akanugan/PF_Validation_PFEGMID_v2_DNNon_vs_DNNon+PR36364/FlatQCD_PU25ns_JetMETJetValidation/slimmedJets/chargedHadronEnergyFraction.pdf
4. chargedHadronEnergyFraction of slimmedJetsAK8 in FlatQCD_PU: https://akanugan.web.cern.ch/akanugan/PF_Validation_PFEGMID_v2_DNNon_vs_DNNon+PR36364/FlatQCD_PU25ns_JetMETJetValidation/slimmedJetsAK8/chargedHadronEnergyFraction.pdf

Base vs Base+PR looks consistent within stat uncertainty.
So I'd say we are fine with this PR.

[slava77]

Base vs Base+PR looks consistent within stat uncertainty.
So I'd say we are fine with this PR.

@swagata87
thank you for the checks and the confirmation.

[slava77]

+reconstruction

for #36364 c0aec63

• jenkins tests show differences in electron track isolation, which is expected to increase (recover to the level seen in CKF tracks)
• additional checks provided in check both algo and originalAlgo in electron track isolation calculators #36364 (comment) confirm that there are no downsides or significant differences

[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. @perrotta, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

[perrotta]

+1