Switch to pfCandidates for b tagging for Puppi jets

[ghost]

Solution to the issue discussed here: #22608

Switch back to standard pfCandidates for b tagging also for Puppi jets, reverting the change introduced in: #18666

This change was introduced as it was giving some b tagging performance improvement for Phase2, which is not there anymore with the new Puppi cuts for Phase 2.

Going back to pfCandidates for b tagging recovers in a satisfactory way the performance as shown in the red curves below. It's also a safer choice as it does not expose b tagging to changes in the Puppi algorithm and its parameters.

The performance of the red curves (standard b tagging with pfCandidates and Phase 2 Puppi cuts) is close now to the black curves, using special custom Phase 2 cuts, which was essentially the attempted solution in #22608.

The PR can be backported to 93X and included in: #22554

[cmsbuild]

The code-checks are being triggered in jenkins.

[ghost]

@kpedro88

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-22696/4073

[cmsbuild]

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

It involves the following packages:

PhysicsTools/PatAlgos

@perrotta, @monttj, @cmsbuild, @slava77, @gpetruc, @arizzi can you please review it and eventually sign? Thanks.
@TaiSakuma, @gouskos, @imarches, @ahinzmann, @acaudron, @mmarionncern, @rappoccio, @jdamgov, @jdolen, @nhanvtran, @gpetruc, @gkasieczka, @schoef, @ferencek, @mverzett, @mariadalfonso, @pvmulder, @seemasharmafnal, @JyothsnaKomaragiri 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

[kpedro88]

assign upgrade

[cmsbuild]

New categories assigned: upgrade

@kpedro88 you have been requested to review this Pull request/Issue and eventually sign? Thanks

[kpedro88]

@imarches since this change only targets phase2, it should not modify the settings for other eras. Please restrict this change to be activated by a phase2 era.

[ghost]

This is on purpose targeting all eras.

The change in #18666 was introduced as, with the Puppi cuts we had at the time, there were some performance gains for Phase 2. No meaningful performance improvement was seen elsewhere and it was extended to all eras just for consistency, as explained in #18426

With the motivation gone for Phase 2 to keep things as in #18666 , there is then no reason to keep that configuration in other eras.

[kpedro88]

Okay, if this is acceptable for BTV then it's the easiest approach.

[kpedro88]

please test

[cmsbuild]

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/27027/console Started: 2018/03/21 19:09

[cmsbuild]

+1
Tested at: d6822ae
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22696/27027/summary.html

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-22696/27027/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 244 differences found in the comparisons
• DQMHistoTests: Total files compared: 29
• DQMHistoTests: Total histograms compared: 2497880
• DQMHistoTests: Total failures: 1
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2497703
• DQMHistoTests: Total skipped: 176
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 1.33000000018 KiB( 21 files compared)
• Checked 118 log files, 9 edm output root files, 29 DQM output files

[kpedro88]

+1
changes limited to slimmedJetsPuppi collections in miniAOD

[slava77]

+1

for #22696 d6822ae

• changes are as described
• jenkins tests pass and comparisons with the baseline show differences only in slimmedJetsPuppi pairDiscriVector data members. The changes have a somewhat common feature with a reduction in the default values of the discriminants. This is expected because we should now be getting more inputs.

Here is one plot for 136.7611 (JetHT Run2016E remini )

It appears that we still do not have any plots for the puppi jets btagging in the standard DQM.

[fabiocos]

+1

[fabiocos]

merge