EGM IDs for MiniAOD Reprocessing (74X)

[lgray]

Contains new IDs for Spring15.

• Cutbased variants for 50ns and 25ns
• MVA base variants, 25ns for electrons, 25ns/50ns for photons.
• HEEPV6

Bools are embedded in the pat objects for these IDs.

Please note that the Photon PF isolation variables are now overwritten in MiniAOD to give the values that are used as input to the cutbased photon ID.
Links to all documentation can be found in the corresponding python configuration files.

HEEP5.1 is removed.

Embedding of ID variables (not the cut flow digests) in PAT is enabled through object modifiers.

• All things needed to recalculate MVA values, as well as mva values and categories themselves.

This PR is dependent on #10556.

Weight files are absent from this PR.

Now includes backport of: #10873

@ikrav @cmkuo
@gpetruc @arizzi

[cmsbuild]

A new Pull Request was created by @lgray (Lindsey Gray) for CMSSW_7_4_X.

EGM IDs for MiniAOD Reprocessing

It involves the following packages:

DataFormats/EgammaCandidates
PhysicsTools/PatAlgos
RecoEgamma/EgammaTools
RecoEgamma/ElectronIdentification
RecoEgamma/PhotonIdentification

@cmsbuild, @cvuosalo, @vadler, @monttj, @slava77 can you please review it and eventually sign? Thanks.
@rappoccio, @Sam-Harper, @imarches, @ahinzmann, @acaudron, @mmarionncern, @TaiSakuma, @jdolen, @nhanvtran, @schoef, @ferencek, @gpetruc, @mariadalfonso, @pvmulder this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
If you are a L2 or a release manager you can ask for tests by saying 'please test' in the first line of a comment.
@Degano you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[cmsbuild]

Pull request #10598 was updated. @cmsbuild, @cvuosalo, @vadler, @monttj, @slava77 can you please check and sign again.

[lgray]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.

[cmsbuild]

+1
Tested at: a9464da
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-10598/7005/summary.html

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-10598/7005/summary.html

[slava77]

@lgray is there a PR in the cms-data for the xml files already?
What is the net effect on the memory use? (e.g., using step3 from 4.77, since it's closer in size to T0 workflows and has some egamma objects)

[lgray]

@slava77 yes, links: cms-data/RecoEgamma-PhotonIdentification#3 cms-data/RecoEgamma-ElectronIdentification#5

[lgray]

@slava77 Net effect on RSS was something like a 40-50MB reduction in RSS. igprof showed something bigger.

[cmsbuild]

-1
Tested at: 0533d42
When I ran the RelVals I found an error in the following worklfows:
4.22 step1

DAS Error

1000.0 step1

DAS Error

1001.0 step1

DAS Error

1003.0 step1

DAS Error

you can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-10598/7556/summary.html

[slava77]

+1

for #10598 0533d42

• changes are consistent with integration of commits made in Important-> EDM IDs update, Fixed typo in effective areas. #10873 on top of previously signed off version of this PR in EGM IDs for MiniAOD Reprocessing (74X) #10598 b4877ad [checked by cherry-picking]
• jenkins tests passed as far as the could with the infrastructure issues.
• tested locally with a short matrix and 200 events from 251721 DoubleMuon and DoubleEG in CMSSW_7_4_10_patch1 /test area sign579/
  • there are no changes in the monitored quantities in DQM or in fwlite reco comparisons
  • compared to 75X/76X, there is a larger (about x2) contribution in CPU from photonIDValueMapProducer: the code is using ptrAt, which comes through a RefToBase in 74X. The PhotonIDValueMapProducer is the largest consumer of edm::Viewreco::Candidate::ptrAt method https://slava77sk.web.cern.ch/slava77sk/reco/cgi-bin/igprof-navigator/CMSSW_7_4_10_patch1-sign579-cvmfs.251721.DoubleEG.pp/985 . This code, as implemented doesn't seem to need the persisted Ptr/Ref info and could operate with plain c reference.
  • AOD and RECO outputs are unchanged
  • MINIAOD output shows mild increase in the size of electrons and photons

-------------------------------------------------------------
 orig, B       new, B      delta, B   delta, %   deltaJob, %    branch 
-------------------------------------------------------------
     2602 ->        2636         34      1.3   0.10     patElectrons_slimmedElectrons__RECO.
     1947 ->        2046         99      5.0   0.28     patPhotons_slimmedPhotons__RECO.

[lgray]

@cmsbuild please test

@monttj can you please sign? Thanks!

[cmsbuild]

The tests are being triggered in jenkins.

[cmsbuild]

+1
Tested at: 0533d42
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-10598/7601/summary.html

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-10598/7601/summary.html

[monttj]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_7_4_X IBs once checked with relvals in the development release cycle of CMSSW (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @Degano, @smuzaffar

[slava77]

@davidlange6 this one is merged in #10649 already