Monitor conversion tracks in tracking validation

[makortel]

This PR adds MultiTrackValidator instance to monitor conversion tracks: currently convStepTracks, conversionStepTracks, ckfInOutTracksFromConversions, and ckfOutInTracksFromConversions collections. Tracks from these collections are matched against "conversion TrackingParticles", i.e. electron/positron TrackingParticles (abs(pdgId) == 11) that have photon TrackingParticles as their parents (similar to PhotonValidator and PhotonMCTruthFinder). A consequence is that a track that is matched to a non-conversion TrackingParticle is classified as a fake.

Plot script will be updated separately (to not to conflict with #12893).

On the same go I modified the seeding+building MTV instances (introduced in #12747) to be harvested only in the tracingOnly mode (where they are present) to silence printouts

%MSG-e DQMGenericClient:  DQMGenericClient:postProcessorTrack@endJob  11-Jan-2016 15:49:48 CET PostEndRun
 DQMGenericClient::findAllSubdirectories ==> Missing folder Tracking/TrackSeeding !!!
%MSG
%MSG-e DQMGenericClient:  DQMGenericClient:postProcessorTrack@endJob  11-Jan-2016 15:49:48 CET PostEndRun
 DQMGenericClient::findAllSubdirectories ==> Missing folder Tracking/TrackBuilding !!!
%MSG

Tested in CMSSW_8_0_X_2016-01-11-1100, no changes expected in existing plots.

@rovere @VinInn @slava77 @matteosan1

[cmsbuild]

A new Pull Request was created by @makortel (Matti Kortelainen) for CMSSW_8_0_X.

It involves the following packages:

CommonTools/RecoAlgos
Validation/Configuration
Validation/RecoTrack

@civanch, @cvuosalo, @mdhildreth, @cmsbuild, @deguio, @slava77, @vanbesien, @davidlange6 can you please review it and eventually sign? Thanks.
@rappoccio, @ahinzmann, @abbiendi, @GiacomoSguazzoni, @jhgoh, @VinInn, @jdolen, @rovere, @wmtford, @istaslis, @cerati, @dgulhan this is something you requested to watch as well.
@slava77, @Degano, @smuzaffar you are the release manager for this.

Following commands in first line of a comment are recognized

• +1|approve[d]|sign[ed]: L1/L2's to approve it
• -1|reject[ed]: L1/L2's to reject it
• assign <category>[,<category>[,...]]: L1/L2's to request signatures from other categories
• unassign <category>[,<category>[,...]]: L1/L2's to remove signatures from other categories
• hold: L1/all L2's/release manager to mark it as on hold
• unhold: L1/user who put this PR on hold
• merge: L1/release managers to merge this request
• [@cmsbuild,] please test: L1/L2 and selected users to start jenkins tests
• [@cmsbuild,] please test with cms-sw/cmsdist#<PR>: L1/L2 and selected users to start jenkins tests using externals from cmsdist

[VinInn]

@cmsbuild, please test

[cmsbuild]

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/10460/console

[cmsbuild]

+1
Tested at: e8aba6a
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-12909/10460/summary.html

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-12909/10460/summary.html

[cvuosalo]

@makortel: There are numerous very tiny differences in some Jenkins DQM plots related to cutsRecoFromPVHp_trackingParticleRecoAsssociation, in workflow 50202.0 for example.

You said no changes are expected. Might these differences indicate a problem?

[makortel]

@cvuosalo All the differences (in 50202.0 and 25202.0) are in _Mean and _Sigma plots produced by DQMGenericClient in the harvesting step, i.e. numerical differences in fits. The histograms used as an input to the fits show no differences.

@slava77 Since we encounter these every now and then, would it make sense to add an option to compareValHists.C to ignore diffBins changes in X_Mean and X_Sigma if X are equal (and e.g. P(KS)=1) and use that in PR tests?

[slava77]

On 1/12/16 12:01 PM, Matti Kortelainen wrote:

@cvuosalo https://github.com/cvuosalo All the differences (in 50202.0
and 25202.0) are in |_Mean| and |_Sigma| plots produced by
DQMGenericClient in the harvesting step, i.e. numerical differences in
fits. The histograms used as an input to the fits show no differences.

@slava77 https://github.com/slava77 Since we encounter these every now
and then, would it make sense to add an option to |compareValHists.C| to
ignore diffBins changes in |X_Mean| and |X_Sigma| if |X| are equal (and
e.g. P(KS)=1) and use that in PR tests?

The script is not meant to be so smart.

Technically, all fits can change within numerical precision if the
fitter has a memory of the past fits.
So, it can spread beyond MTV plots.

—
Reply to this email directly or view it on GitHub
#12909 (comment).

[cvuosalo]

+1

For #12909 e8aba6a

Adding monitoring of conversion tracks in tracking validation.

The code changes are satisfactory, and Jenkins tests against baseline CMSSW_8_0_X_2016-01-11-1100 show no significant differences, as expected. The Jenkins DQM plots do show tiny, insignificant differences that are probably due to minor numerical differences in fits, as discussed above.

[civanch]

+1

[makortel]

Could DQM (@deguio, @vanbesien) please review and sign? Thanks.

[deguio]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_8_0_X IBs (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @slava77, @davidlange6, @Degano, @smuzaffar

[davidlange6]

+1