PUPPI MET in RECO #29254

ahinzmann · 2020-03-20T16:05:22Z

PR description:

This PR computes puppiNoLep weights and puppi MET in RECO and stores the puppiNoLep weight as a value map. PUPPI MET is clustered based on standard PF candidates together with the value map of weights. A new boolean "isWeighted" is added to the reco::PFMet interface to indicate whether weights were applied during jet clustering. The packed candidates pick up the value map such that puppi does not need to be re-computed in MiniAOD.

The same was done for the puppi and PUPPI jets in #28936.

Other changes:

Produce puppiForMET directly in PuppiProducer instead of a two step procedure puppiNoLep->puppiForMET and rename it to puppiNoLep, consistent with the name in PackedCandidates
Implement fillDescriptions for PFMetProducer
Rename PFMET_cfi to pfMet_cfi everywhere (for consistency with the module name)
Fix input jet collection of pfMetEI
Remove pfMetEI from RecoMET_EventContent (however, leave EITopPAG_EventContent_cff untouched)
Resolve issue cleanup miniAOD workflow after moving PUPPI to reco #29264
By moving to the same input PF candidate collection for puppi MET and puppi jets (with just different weights), PUPPI MET significance and unclustered pT computation then don't rely on dR matching as implemented in Fix MET unclustered energy computation #29331 anymore.
Store ref to PF candidate inside pat muons (just like done for all other PAT objects). PUPPI MET significance and unclustered pT computation then don't rely on dR matching as implemented in Fix MET Significance, Covariance Matrix and unclustered energy lepton subtraction #29385 anymore.
Fix particle multiplicity computation in xy-correction for PUPPI MET. Previously, particles with a weight of 0 could also count in the multiplicity as a function of which the xy-corrections are computed. Since xy-correction are not needed/recommended for PUPPI MET, this fix is not backported.
For HI workflows, raise candidate pT threshold for chsMet and trkMet to 999. consistent with pfMet.

PR validation:

runTheMatrix -l limited

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

No backport foreseen.

cmsbuild · 2020-03-20T16:05:50Z

The code-checks are being triggered in jenkins.

ahinzmann · 2020-03-20T16:06:16Z

please wait before testing, I hit enter too quickly

cmsbuild · 2020-03-20T16:13:54Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-29254/14294

This PR adds an extra 92KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File CommonTools/PileupAlgos/plugins/PuppiProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File CommonTools/PileupAlgos/plugins/PuppiProducer.h modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/plugins/PATPackedCandidateProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/python/slimming/miniAOD_tools.py modified in PR(s): PUPPI jets in RECO #28936, New Tau ID data format #28417
- File PhysicsTools/PatUtils/python/tools/runMETCorrectionsAndUncertainties.py modified in PR(s): PUPPI jets in RECO #28936

cmsbuild · 2020-03-20T16:14:15Z

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

It involves the following packages:

CommonTools/ParticleFlow
CommonTools/PileupAlgos
DataFormats/METReco
PhysicsTools/PatAlgos
PhysicsTools/PatUtils
RecoMET/Configuration
RecoMET/METAlgorithms
RecoMET/METProducers

@perrotta, @cmsbuild, @santocch, @slava77 can you please review it and eventually sign? Thanks.
@rappoccio, @gouskos, @hatakeyamak, @emilbols, @peruzzim, @seemasharmafnal, @mmarionncern, @smoortga, @jdolen, @ferencek, @rovere, @jdamgov, @nhanvtran, @gkasieczka, @schoef, @andrzejnovak, @clelange, @riga, @JyothsnaKomaragiri, @mverzett, @cbernet, @gpetruc, @mariadalfonso this is something you requested to watch as well.
@davidlange6, @silviodonato, @fabiocos you are the release manager for this.

cms-bot commands are listed here

cmsbuild · 2020-03-20T19:18:51Z

The code-checks are being triggered in jenkins.

slava77 · 2020-03-20T19:20:43Z

@ahinzmann
will this need to be tested together with #28936 ?
Please clarify.
Thank you.

cmsbuild · 2020-03-20T19:25:03Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-29254/14301

This PR adds an extra 92KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File CommonTools/PileupAlgos/plugins/PuppiProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File CommonTools/PileupAlgos/plugins/PuppiProducer.h modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/plugins/PATPackedCandidateProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/python/slimming/miniAOD_tools.py modified in PR(s): PUPPI jets in RECO #28936, New Tau ID data format #28417
- File PhysicsTools/PatUtils/python/tools/runMETCorrectionsAndUncertainties.py modified in PR(s): PUPPI jets in RECO #28936

cmsbuild · 2020-03-20T19:25:31Z

Pull request #29254 was updated. @perrotta, @cmsbuild, @santocch, @slava77 can you please check and sign again.

ahinzmann · 2020-03-20T19:27:27Z

This can be tested standalone. Once #28936 is merged in CMSSW, I can do the merge here as well. I'm still finishing one round of local testing. Once they finish, it would be good to test here.

cmsbuild · 2020-03-20T22:06:43Z

The code-checks are being triggered in jenkins.

ahinzmann · 2020-03-20T22:12:07Z

local test ran successful. Could this version be tested?

cmsbuild · 2020-03-20T22:18:38Z

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-29254/14303

This PR adds an extra 96KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File CommonTools/PileupAlgos/plugins/PuppiProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File CommonTools/PileupAlgos/plugins/PuppiProducer.h modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/plugins/PATPackedCandidateProducer.cc modified in PR(s): PUPPI jets in RECO #28936
- File PhysicsTools/PatAlgos/python/slimming/miniAOD_tools.py modified in PR(s): PUPPI jets in RECO #28936, New Tau ID data format #28417
- File PhysicsTools/PatUtils/python/tools/runMETCorrectionsAndUncertainties.py modified in PR(s): PUPPI jets in RECO #28936

Code check has found code style and quality issues which could be resolved by applying following patch(s)

code-format:
https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-29254/14303/code-format.patch
e.g. curl https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-29254/14303/code-format.patch | patch -p1
You can also run scram build code-format to apply code format directly

jfernan2 · 2020-04-29T08:54:05Z

+1

slava77 · 2020-04-29T23:35:45Z

+1

for #29254 d6e1687

code changes are in line with the PR description and the follow up review
jenkins tests pass and comparisons with the baseline show differences in expected places in puppi MET related variables (see earlier discussion)

resolves #29264

silviodonato · 2020-05-04T09:00:00Z

Do you have comments @santocch ?

slava77 · 2020-05-05T13:08:39Z

@silviodonato
please include this in pre7

silviodonato · 2020-05-05T17:41:18Z

merge

silviodonato · 2020-05-06T06:10:40Z

We are getting some errors in the IB. Please look at #29744

santocch · 2020-05-10T21:24:35Z

+1

cmsbuild · 2020-05-10T21:24:59Z

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 be automatically merged.

slava77 · 2020-08-05T23:59:02Z

PhysicsTools/PatUtils/python/tools/runMETCorrectionsAndUncertainties.py

            if not self.getvalue("onMiniAOD"):
-              pfCandsNoJetsNoEle.useDeltaRforFootprint = True
              pfCandsNoJetsNoEle.veto = "pfeGammaToCandidate:electrons"
            addToProcessAndTask("pfCandsNoJetsNoEle"+postfix, pfCandsNoJetsNoEle, process, task)


I was debugging differences in UL re-miniAOD between 106X and 112X, and was puzzled by the somewhat small but numerous changes in slimmedMETsNoHF UnclusteredEnUp/Down.
If I understood correctly (although I did not add any debug printouts to confirm), the problem is that the "NoHF" is based on noHFCands, which are copies (not Ptrs) of the original PF cands

cmssw/CommonTools/ParticleFlow/interface/GenericPFCandidateSelectorDefinition.h

Lines 31 to 34 in 3402e09

if (selector_(*pfc)) {

selected_.push_back(reco::PFCandidate(*pfc));

reco::PFCandidatePtr ptrToMother(hc, key);

selected_.back().setSourceCandidatePtr(ptrToMother);

So, the CandPtrProjector can not match the vetoing electron's sourceCandidatePtr (the original particleFlow) to the noHFCands (another check by the cand's sourceCandidatePtr may suffice though).
So, either we need an extension in CandPtrProjector to also check for the main collection sourceCandidatePtr or instead have useDeltaRforFootprint = True

I'll re-post this as an issue.

cmsbuild added this to the CMSSW_11_1_X milestone Mar 20, 2020

cmsbuild added analysis-pending code-checks-pending comparison-pending orp-pending pending-signatures reconstruction-pending tests-pending labels Mar 20, 2020

cmsbuild added code-checks-approved and removed code-checks-pending labels Mar 20, 2020

ahinzmann mentioned this pull request Mar 20, 2020

PUPPI jets in RECO #28936

Merged

cmsbuild added code-checks-pending and removed code-checks-approved labels Mar 20, 2020

cmsbuild added code-checks-approved and removed code-checks-pending labels Mar 20, 2020

cmsbuild added code-checks-pending dqm-pending and removed code-checks-approved labels Mar 20, 2020

cmsbuild added dqm-approved and removed dqm-pending labels Apr 29, 2020

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

cmsbuild mentioned this pull request Apr 30, 2020

PUPPI tune v14 (phase-2 excluded) #29600

Merged

slava77 mentioned this pull request May 1, 2020

cleanup miniAOD workflow after moving PUPPI to reco #29264

Closed

cmsbuild added orp-approved and removed orp-pending labels May 5, 2020

cmsbuild merged commit 6d0e276 into cms-sw:master May 5, 2020

This was referenced May 6, 2020

Segmentation fault in DQMStore #29743

Closed

Exception: Original collection and shifted collection are of different size in ShiftedParticleMETcorrInputProducer #29744

Closed

silviodonato mentioned this pull request May 6, 2020

Revert "PUPPI MET in RECO" #29748

Merged

slava77 mentioned this pull request May 7, 2020

run2 re-miniAOD setup based on UL inputs #29756

Merged

silviodonato mentioned this pull request May 7, 2020

PUPPI MET in RECO #29768

Closed

ahinzmann mentioned this pull request May 7, 2020

PUPPI MET in RECO #29769

Merged

cmsbuild added analysis-approved fully-signed and removed analysis-pending pending-signatures labels May 10, 2020

slava77 reviewed Aug 5, 2020

View reviewed changes

slava77 mentioned this pull request Aug 6, 2020

migrate from run2_miniAOD_devel to run2_miniAOD_UL for miniAOD setup UL workflows #31072

Merged

perrotta mentioned this pull request Nov 12, 2020

Backport #31927 (PuppiMET with PUPPIv15 and UL17 PUID) #32076

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

PUPPI MET in RECO #29254

PUPPI MET in RECO #29254

ahinzmann commented Mar 20, 2020 •

edited

Loading

cmsbuild commented Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

slava77 commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

jfernan2 commented Apr 29, 2020

slava77 commented Apr 29, 2020

silviodonato commented May 4, 2020

slava77 commented May 5, 2020

silviodonato commented May 5, 2020

silviodonato commented May 6, 2020

santocch commented May 10, 2020

cmsbuild commented May 10, 2020

slava77 Aug 5, 2020

	if (selector_(*pfc)) {
	selected_.push_back(reco::PFCandidate(*pfc));
	reco::PFCandidatePtr ptrToMother(hc, key);
	selected_.back().setSourceCandidatePtr(ptrToMother);

PUPPI MET in RECO #29254

PUPPI MET in RECO #29254

Conversation

ahinzmann commented Mar 20, 2020 • edited Loading

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 Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

slava77 commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

ahinzmann commented Mar 20, 2020

cmsbuild commented Mar 20, 2020

jfernan2 commented Apr 29, 2020

slava77 commented Apr 29, 2020

silviodonato commented May 4, 2020

slava77 commented May 5, 2020

silviodonato commented May 5, 2020

silviodonato commented May 6, 2020

santocch commented May 10, 2020

cmsbuild commented May 10, 2020

slava77 Aug 5, 2020

Choose a reason for hiding this comment

ahinzmann commented Mar 20, 2020 •

edited

Loading