TICL reconstruction update

[rovere]

That's a backport of #31906

It's not exactly a perfect backport, but it introduces the same functionalities and improvements, together with the needed components.

Please refer to that PR to see what has been changed and what to expect from this PR, too.

[cmsbuild]

A new Pull Request was created by @rovere (Marco Rovere) for CMSSW_11_1_X.

It involves the following packages:

DataFormats/HGCalReco
RecoEgamma/EgammaTools
RecoHGCal/Configuration
RecoHGCal/TICL
RecoLocalCalo/HGCalRecAlgos
RecoLocalCalo/HGCalRecProducers
RecoParticleFlow/PFClusterProducer
Validation/HGCalValidation

@perrotta, @andrius-k, @kmaeshima, @ErnestaP, @kpedro88, @cmsbuild, @jfernan2, @fioriNTU, @slava77, @jpata can you please review it and eventually sign? Thanks.
@felicepantaleo, @jainshilpi, @argiro, @bsunanda, @pfs, @thomreis, @varuns23, @seemasharmafnal, @mmarionncern, @sethzenz, @lgray, @apsallid, @vandreev11, @Sam-Harper, @cbernet, @cseez, @deguio, @hatakeyamak, @clelange, @rchatter, @riga, @edjtscott, @sobhatta, @lecriste, @afiqaize this is something you requested to watch as well.
@silviodonato, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

[rovere]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.

• CMSSW_11_1_X_2020-10-22-1100/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/10219/console Started: 2020/10/22 19:26

[cmsbuild]

+1
Tested at: aa350cc
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-f9b338/10219/summary.html
CMSSW: CMSSW_11_1_X_2020-10-22-1100
SCRAM_ARCH: slc7_amd64_gcc820

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-f9b338/10219/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 2552 differences found in the comparisons
• DQMHistoTests: Total files compared: 36
• DQMHistoTests: Total histograms compared: 2784776
• DQMHistoTests: Total failures: 6002
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2778724
• DQMHistoTests: Total skipped: 50
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 35 files compared)
• Checked 152 log files, 16 edm output root files, 36 DQM output files

[jfernan2]

@rovere I understand we should not expect the same changes as in the master #31906 since they are not identical, indeed the number of changes is quite different.
However in both cases plots in HGCAL / TICLPFCandidates / 3 get now zero entries and HGCAL / HGCalValidator / ticlMultiClustersFromTrackstersMerge/Trk now get many Denominator null values
https://tinyurl.com/y3zx7cww
https://tinyurl.com/yyrcgaoe

I guess this is just a matter of a new definition, but I wonder if it is expected. I will post this same question in master PR

[cmsbuild]

Pull request #31907 was updated. @perrotta, @andrius-k, @kmaeshima, @ErnestaP, @kpedro88, @cmsbuild, @jfernan2, @fioriNTU, @slava77, @jpata can you please check and sign again.

[rovere]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.

• CMSSW_11_1_X_2020-10-30-1100/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/10420/console Started: 2020/10/30 23:43

[cmsbuild]

+1
Tested at: de99de0
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-f9b338/10893/summary.html
CMSSW: CMSSW_11_1_X_2020-11-20-1100
SCRAM_ARCH: slc7_amd64_gcc820

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-f9b338/10893/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 3943 differences found in the comparisons
• DQMHistoTests: Total files compared: 36
• DQMHistoTests: Total histograms compared: 2784828
• DQMHistoTests: Total failures: 13638
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2771140
• DQMHistoTests: Total skipped: 50
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 35 files compared)
• Checked 152 log files, 28 edm output root files, 36 DQM output files

[perrotta]

+1

• Backport of HFNose: add ticl iteration #29837, [bugfix] HFNose in TICL - Move template instantiation to .cc #30007 RecoHGCal/TICL: optimize calculation for comparison in HGCDoublet::areAligned #30815 added esConsumes to modules in RecoHGCal/TICL and related packages #31013 (partial) Don't use uninitialized edm::Handle in TrackstersProducer #31031 [HGCAL] CLUE and regional factors #31730 drop type specs in RecoHGCal #31881 [HGCal] TICL v3 major upgrade #31906 [HGCAL] Corrections in noise vector in CLUE #32021 HGCAL: restore distance cut in CLUE position calculation #32022, plus a few additional TICL related improvements/cleanings already merged in the master; in most cases the TICL code is now aligned to the one currently in the CMSSW master
• Differences in Phase2 outputs wrt the baseline are as expected due to the backported PRs. It was decided at the ORP that the "no change" rule for closed releases can be relaxed now for 11_1_X, trying to have the most updated reconstruction for the forthcoming Phase2 HLT TDR studies
• New objects saved (ticlTrackstersMerge, ticlTrackstersTrkEM, ticlMultiClustersFromTrackstersTrkEM, ticlTrackstersHFNoseEM, ticlTrackstersHFNoseMIP, ticlTrackstersHFNoseMerge), others removed (ticlCandidateFromTracksters)

[jfernan2]

+1

[kpedro88]

+upgrade

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_11_1_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_11_2_X is complete. This pull request will now be reviewed by the release team before it's merged. @silviodonato, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

[qliphy]

+1

[hatakeyamak]

@rovere I see that this is now integrated to 11_1.
I was just wondering: does this run with injectTICLintoPF without a modification to ChargedHadronPFTrackIsolationProducer (e.g. backport of #32202) or do we need this backport?