Adapt track-vertex association in PUPPI (v15) [11_1_X] #31494

ahinzmann · 2020-09-17T13:10:00Z

PR description:

The purpose of this backport is to enable study of PUPPI tune v15 in the context of the HLT-TDR. The default configuration is not touched here, but may only be changed after successful validation of PUPPI v15 for Phase-2.
This introduces the option to use a new treatment of the track-vertex association for charged particles in PUPPI. The new version is called v15.
The vertex association is adapted to deal with split vertices in high hadronic activity events (more on this here: https://indico.cern.ch/event/944339/contributions/3968152/attachments/2086414/3505127/splitverticesandhighptjets.pdf).
Charged particles associated to one of the leading two pileup vertices by the vertex fit (fromPV=0), but found within dz<0.2 of the primary vertex are treated as coming from the primary vertex.
This reduces MET tails due to particles that are wrongly associated to a PU vertex, when the PV is reconstructed as 2-3 split vertices.
The overall impact on jet response, jet efficiency, fake rate, boosted object tagging is negligible.
Thus JEC and SFs derived with PUPPI v14 are applicable to PUPPI v15.
A summary of the studies is in:
https://indico.cern.ch/event/942994/contributions/3978585/attachments/2088015/3508036/puppiv15summary-1.pdf
An additional check on softdrop mass response for high pT AK8 jets can be found here:
https://indico.cern.ch/event/942994/#7-virtual-impact-of-puppi-tune

PR validation:

scram b run-tests
runTheMatrix.py -l limited

if this PR is a backport please specify the original PR and why you need to backport that PR:

Backport of #31174 and #31887

cmsbuild · 2020-09-17T13:10:25Z

A new Pull Request was created by @ahinzmann for CMSSW_11_1_X.

It involves the following packages:

CommonTools/PileupAlgos

@perrotta, @jpata, @cmsbuild, @santocch, @slava77 can you please review it and eventually sign? Thanks.
@rappoccio, @riga, @jdolen, @gkasieczka, @hatakeyamak, @clelange 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

Backported from Adapt track-vertex association in PUPPI (v15) #31174

slava77 · 2020-09-17T13:15:54Z

@cmsbuild please test

cmsbuild · 2020-09-17T13:16:12Z

The tests are being triggered in jenkins.

CMSSW_11_1_X_2020-09-17-1100/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/9404/console Started: 2020/09/17 15:17

cmsbuild · 2020-09-17T14:43:43Z

+1
Tested at: 9e8a190
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-8df54e/9404/summary.html
CMSSW: CMSSW_11_1_X_2020-09-17-1100
SCRAM_ARCH: slc7_amd64_gcc820

cmsbuild · 2020-09-17T14:43:52Z

Comparison job queued.

cmsbuild · 2020-09-17T16:30:50Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-8df54e/9404/summary.html

@slava77 comparisons for the following workflows were not done due to missing matrix map:

/data/cmsbld/jenkins/workspace/compare-root-files-short-matrix/data/PR-8df54e/23234.1001_TTbar_14TeV+RecoFullGlobal_TestOldDigi_2026D49+HARVESTFullGlobal_2026D49

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 4 differences found in the comparisons
DQMHistoTests: Total files compared: 36
DQMHistoTests: Total histograms compared: 2780792
DQMHistoTests: Total failures: 3
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 2780739
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

qliphy · 2020-10-27T01:05:21Z

please test

cmsbuild · 2020-10-27T01:05:47Z

The tests are being triggered in jenkins.

slava77 · 2020-10-28T14:38:08Z

@cmsbuild please test

it looks like the last set of tests got lost

cmsbuild · 2020-10-28T14:38:28Z

The tests are being triggered in jenkins.

CMSSW_11_1_X_2020-10-28-1100/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/10349/console Started: 2020/10/28 15:39

cmsbuild · 2020-10-28T16:04:21Z

+1
Tested at: b4d614f
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-8df54e/10349/summary.html
CMSSW: CMSSW_11_1_X_2020-10-28-1100
SCRAM_ARCH: slc7_amd64_gcc820

cmsbuild · 2020-10-28T16:04:23Z

Comparison job queued.

cmsbuild · 2020-10-28T18:04:01Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-8df54e/10349/summary.html

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 36
DQMHistoTests: Total histograms compared: 2784776
DQMHistoTests: Total failures: 1
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 2784725
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

slava77 · 2020-10-29T22:27:30Z

+1

for #31494 b4d614f

the update is consistent with a backport of Adapt track-vertex association in PUPPI (v15) #31174 (the original v15) and Sync puppi tune for phase2 with run3 #31887 (the phase-2 update for it); changes are made in the confines of the no-change policy
jenkins tests pass and comparisons with the baseline show no differences

However,
please note that in 11_1_X the no-change policy is relaxed: changes are allowed for phase2.
If the code is needed for HLT TDR related studies, it may be practical to put the relevant phase2.toModify updates in CommonTools/PileupAlgos/python/Puppi_cff.py instead of CommonTools/PileupAlgos/python/customizePuppiTune_cff.py
IIRC, the pre-phase2 setup was not explicitly granted free changes/updates.
So, the customizePuppiTune_cff will likely have to stay for Run3 or earlier Run2 studies/samples.

santocch · 2020-11-03T17:45:57Z

+1

cmsbuild · 2020-11-03T17:46:17Z

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)

silviodonato · 2020-11-04T09:57:30Z

+1

ahinzmann added 7 commits September 17, 2020 09:24

adapt to puppi v15

89a01ac

comments from Slava

81cb3f2

code format

6928102

no change policy

d356d0c

update v15 recipe

d7bbf8c

undo diff

a428eb1

undo diff

9e8a190

cmsbuild added this to the CMSSW_11_1_X milestone Sep 17, 2020

cmsbuild added analysis-pending comparison-pending orp-pending pending-signatures reconstruction-pending tests-pending labels Sep 17, 2020

cmsbuild added tests-started and removed tests-pending labels Sep 17, 2020

cmsbuild added tests-approved and removed tests-started labels Sep 17, 2020

cmsbuild added comparison-available and removed comparison-pending labels Sep 17, 2020

adapt for phase2

d8e6f80

cmsbuild added comparison-pending and removed comparison-available tests-approved labels Sep 17, 2020

cmsbuild added tests-started and removed tests-pending labels Oct 27, 2020

cmsbuild added tests-pending tests-started and removed tests-started tests-pending labels Oct 28, 2020

cmsbuild added tests-approved and removed tests-started labels Oct 28, 2020

cmsbuild added comparison-available and removed comparison-pending labels Oct 28, 2020

cmsbuild added reconstruction-approved and removed reconstruction-pending labels Oct 29, 2020

cmsbuild added analysis-approved fully-signed and removed analysis-pending pending-signatures labels Nov 3, 2020

cmsbuild added orp-approved and removed orp-pending labels Nov 4, 2020

cmsbuild merged commit f58146d into cms-sw:CMSSW_11_1_X Nov 4, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adapt track-vertex association in PUPPI (v15) [11_1_X] #31494

Adapt track-vertex association in PUPPI (v15) [11_1_X] #31494

ahinzmann commented Sep 17, 2020 •

edited

cmsbuild commented Sep 17, 2020 •

edited

slava77 commented Sep 17, 2020

cmsbuild commented Sep 17, 2020 •

edited

cmsbuild commented Sep 17, 2020

cmsbuild commented Sep 17, 2020

cmsbuild commented Sep 17, 2020

qliphy commented Oct 27, 2020

cmsbuild commented Oct 27, 2020

slava77 commented Oct 28, 2020

cmsbuild commented Oct 28, 2020 •

edited

cmsbuild commented Oct 28, 2020

cmsbuild commented Oct 28, 2020

cmsbuild commented Oct 28, 2020

slava77 commented Oct 29, 2020

santocch commented Nov 3, 2020

cmsbuild commented Nov 3, 2020

silviodonato commented Nov 4, 2020

Adapt track-vertex association in PUPPI (v15) [11_1_X] #31494

Adapt track-vertex association in PUPPI (v15) [11_1_X] #31494

Conversation

ahinzmann commented Sep 17, 2020 • edited

PR description:

PR validation:

if this PR is a backport please specify the original PR and why you need to backport that PR:

cmsbuild commented Sep 17, 2020 • edited

slava77 commented Sep 17, 2020

cmsbuild commented Sep 17, 2020 • edited

cmsbuild commented Sep 17, 2020

cmsbuild commented Sep 17, 2020

cmsbuild commented Sep 17, 2020

qliphy commented Oct 27, 2020

cmsbuild commented Oct 27, 2020

slava77 commented Oct 28, 2020

cmsbuild commented Oct 28, 2020 • edited

cmsbuild commented Oct 28, 2020

cmsbuild commented Oct 28, 2020

cmsbuild commented Oct 28, 2020

slava77 commented Oct 29, 2020

santocch commented Nov 3, 2020

cmsbuild commented Nov 3, 2020

silviodonato commented Nov 4, 2020

ahinzmann commented Sep 17, 2020 •

edited

cmsbuild commented Sep 17, 2020 •

edited

cmsbuild commented Sep 17, 2020 •

edited

cmsbuild commented Oct 28, 2020 •

edited