Update mkFit for 12_1_0_pre5 #35652

mmasciov · 2021-10-13T12:35:12Z

PR description:

This PR follows #35492 and updates mkFit in view of CMSSW_12_1_0_pre5.

In detail, this PR:

adds the beamspot information to mkFit;
fixes naming of parameters in mkFit README;
(via Update mkFit to V3.4.0-0 cmsdist#7387 and Update mkFit JSON files for CMSSW_12_1_0_pre5 cms-data/RecoTracker-MkFit#6) updates mkFit configuration for detachedTripletStep and pixelLessStep tracking iterations, with a gain in track reconstruction efficiency.

It requires cms-sw/cmsdist#7387 and cms-data/RecoTracker-MkFit#6

PR validation:

In TTbar events with =50:

mkFit (enabled in initialStepPreSplitting, initialStep, highPtTripletStep, detachedQuadStep), before and after this PR: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_PR35652_mkFit-4iter_vsMkFit-4iter/
--> Performance is unchanged (as expected).
mkFit enabled in initialStepPreSplitting, initialStep, highPtTripletStep, detachedQuadStep, detachedTripletStep, pixelLessStep (6-iter) vs. default (4-iter: initialStepPreSplitting, initialStep, highPtTripletStep, detachedQuadStep): http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_PR35652_mkFit-6iter_vsMkFit-4iter/
--> Efficiency for detachedTripletStep is improved (with further reduction in fake rate): http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_PR35652_mkFit-6iter_vsMkFit-4iter/plots_building_detachedTripletStep/effandfakePtEtaPhi.pdf
--> Efficiency for pixelLessStep is improved: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_PR35652_mkFit-6iter_vsMkFit-4iter/plots_building_pixelLessStep/effandfakePtEtaPhi.pdf
--> Timing performance is ~ unchanged: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_PR35652_mkFit-6iter_vsMkFit-4iter/plots_timing.html

For history/completeness, comparisons wrt. standard CKF for all tracking iterations follow:

mkFit vs. CKF performance when mkFit is enabled in initialStepPreSplitting, initialStep, highPtTripletStep, detachedQuadStep, before and after this PR: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_mkFit-4iter_vsCKF_Oct13/
--> Performance is unchanged (as expected).
mkFit vs. CKF performance when mkFit is enabled in initialStepPreSplitting, initialStep, highPtTripletStep, detachedQuadStep, detachedTripletStep, pixelLessStep: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_mkFit-6iter_Oct13_vsCKF/
--> Efficiency for detachedTripletStep is improved (with further reduction in fake rate): http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_mkFit-6iter_Oct13_vsCKF/plots_building_detachedTripletStep/effandfakePtEtaPhi.pdf
--> Efficiency for pixelLessStep is improved: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_mkFit-6iter_Oct13_vsCKF/plots_building_pixelLessStep/effandfakePtEtaPhi.pdf
--> Timing performance is ~ unchanged: http://uaf-10.t2.ucsd.edu/~mmasciov/MkFit_devs/PRvalidation_Oct13/MTV_mkFit-6iter_Oct13_vsCKF/plots_timing.html

cmsbuild · 2021-10-13T12:41:45Z

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35652/25920

ERROR: Build errors found during clang-tidy run.

RecoTracker/MkFit/plugins/MkFitEventOfHitsProducer.cc:180:16: error: no member named 'SetBeamSpot' in 'mkfit::EventOfHits' [clang-diagnostic-error]
  eventOfHits->SetBeamSpot({.x = float(bs.x0()),
               ^
Suppressed 1399 warnings (1398 in non-user code, 1 with check filters).
--
gmake: *** [config/SCRAM/GMake/Makefile.coderules:129: code-checks] Error 2
gmake: *** [There are compilation/build errors. Please see the detail log above.] Error 2

slava77 · 2021-10-13T13:07:08Z

eventOfHits->SetBeamSpot({.x = float(bs.x0()),

@makortel @Dr15Jones
is there a problem to use the aggregate initialization in this case?

makortel · 2021-10-13T14:12:12Z

eventOfHits->SetBeamSpot({.x = float(bs.x0()),

@makortel @Dr15Jones is there a problem to use the aggregate initialization in this case?

Are you asking in a technical sense or in a policy (core rule) sense? (although I don't really see a problem in either case)

slava77 · 2021-10-13T14:29:16Z

Are you asking in a technical sense or in a policy (core rule) sense? (although I don't really see a problem in either case)

well, if there is no problem, why is there an error?
Is there something missing on the code-checks side that does not allow this syntax?

makortel · 2021-10-13T14:34:15Z

To me it seems the error is no member named 'SetBeamSpot' in 'mkfit::EventOfHits' and not the aggregate initialization, but I didn't check the full log.

makortel · 2021-10-13T14:58:12Z

Just to summarize here, the mkfit BeamSpot class has a user-defined constructor and is therefore not an aggregate. I guess the designated initializer working with gcc (with C++17!) is a non-standard extension or something.

cmsbuild · 2021-10-13T15:10:19Z

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35652/25927

ERROR: Build errors found during clang-tidy run.

RecoTracker/MkFit/plugins/MkFitEventOfHitsProducer.cc:180:16: error: no member named 'SetBeamSpot' in 'mkfit::EventOfHits' [clang-diagnostic-error]
  eventOfHits->SetBeamSpot(
               ^
RecoTracker/MkFit/plugins/MkFitEventOfHitsProducer.cc:181:14: error: no member named 'BeamSpot' in namespace 'mkfit' [clang-diagnostic-error]
      mkfit::BeamSpot(bs.x0(), bs.y0(), bs.z0(), bs.sigmaZ(), bs.BeamWidthX(), bs.BeamWidthY(), bs.dxdz(), bs.dydz()));
             ^
Suppressed 1399 warnings (1398 in non-user code, 1 with check filters).
--
gmake: *** [config/SCRAM/GMake/Makefile.coderules:129: code-checks] Error 2
gmake: *** [There are compilation/build errors. Please see the detail log above.] Error 2

cmsbuild · 2021-10-13T17:28:58Z

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35652/25935

ERROR: Build errors found during clang-tidy run.

RecoTracker/MkFit/plugins/MkFitEventOfHitsProducer.cc:180:16: error: no member named 'SetBeamSpot' in 'mkfit::EventOfHits' [clang-diagnostic-error]
  eventOfHits->SetBeamSpot(
               ^
RecoTracker/MkFit/plugins/MkFitEventOfHitsProducer.cc:181:14: error: no member named 'BeamSpot' in namespace 'mkfit' [clang-diagnostic-error]
      mkfit::BeamSpot(bs.x0(), bs.y0(), bs.z0(), bs.sigmaZ(), bs.BeamWidthX(), bs.BeamWidthY(), bs.dxdz(), bs.dydz()));
             ^
Suppressed 1399 warnings (1398 in non-user code, 1 with check filters).
--
gmake: *** [config/SCRAM/GMake/Makefile.coderules:129: code-checks] Error 2
gmake: *** [There are compilation/build errors. Please see the detail log above.] Error 2

makortel · 2021-10-13T17:33:41Z

code-checks with cms.week0.PR_d0bce566/52.0-60549143f91cbd41608af2b631539e4b

cmsbuild · 2021-10-13T17:41:02Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-35652/25936

This PR adds an extra 16KB to repository

cmsbuild · 2021-10-13T17:41:23Z

A new Pull Request was created by @mmasciov (Mario Masciovecchio) for master.

It involves the following packages:

RecoTracker/MkFit (reconstruction)

@jpata, @cmsbuild, @slava77 can you please review it and eventually sign? Thanks.
@makortel, @felicepantaleo, @GiacomoSguazzoni, @JanFSchulte, @rovere, @VinInn, @ebrondol, @gpetruc, @mmusich, @mtosi, @dgulhan this is something you requested to watch as well.
@perrotta, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

makortel · 2021-10-13T17:44:16Z

@cmsbuild, please test with cms-sw/cmsdist#7387, cms-data/RecoTracker-MkFit#6

cmsbuild · 2021-10-13T21:21:41Z

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0b5956/19611/summary.html
COMMIT: b32639d
CMSSW: CMSSW_12_1_X_2021-10-13-1100/slc7_amd64_gcc900
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/35652/19611/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

No significant changes to the logs found
Reco comparison results: 4 differences found in the comparisons
DQMHistoTests: Total files compared: 40
DQMHistoTests: Total histograms compared: 2796791
DQMHistoTests: Total failures: 5
DQMHistoTests: Total nulls: 1
DQMHistoTests: Total successes: 2796763
DQMHistoTests: Total skipped: 22
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: -0.004 KiB( 39 files compared)
DQMHistoSizes: changed ( 312.0 ): -0.004 KiB MessageLogger/Warnings
Checked 169 log files, 37 edm output root files, 40 DQM output files
TriggerResults: no differences found

jpata · 2021-10-15T12:06:09Z

+reconstruction

no reco changes in standard workflows
requires Update mkFit JSON files for CMSSW_12_1_0_pre5 cms-data/RecoTracker-MkFit#6, Update mkFit to V3.4.0-0 cmsdist#7387

cmsbuild · 2021-10-15T12:06:34Z

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 now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

perrotta · 2021-10-19T15:27:37Z

+1

To be merged together with Update mkFit JSON files for CMSSW_12_1_0_pre5 cms-data/RecoTracker-MkFit#6, and Update mkFit to V3.4.0-0 cmsdist#7387

mmasciov added 2 commits October 13, 2021 04:49

Update names of (two) parameters in mkFit README

1a72fd4

Set beamspot in mkFit eventOfHits

e400333

cmsbuild added this to the CMSSW_12_1_X milestone Oct 13, 2021

cmsbuild added code-checks-pending orp-pending pending-signatures reconstruction-pending tests-pending labels Oct 13, 2021

mmasciov mentioned this pull request Oct 13, 2021

Update mkFit JSON files for CMSSW_12_1_0_pre5 cms-data/RecoTracker-MkFit#6

Merged

cmsbuild added code-checks-rejected and removed code-checks-pending labels Oct 13, 2021

mmasciov changed the title ~~Update mk fit 12 1 0 pre5~~ Update mkFit for 12_1_0_pre5 Oct 13, 2021

makortel mentioned this pull request Oct 13, 2021

Update mkFit to V3.4.0-0 cms-sw/cmsdist#7387

Merged

Use mkfit::BeamSpot constructor

ed3032c

cmsbuild added code-checks-pending and removed code-checks-rejected labels Oct 13, 2021

cmsbuild added code-checks-rejected and removed code-checks-pending labels Oct 13, 2021

Override beamspot parameter in customizeHLTIter0ToMkFit.py

b32639d

cmsbuild added code-checks-pending and removed code-checks-rejected labels Oct 13, 2021

mmasciov mentioned this pull request Oct 13, 2021

Backport of PR 35652 (++) trackreco/cmssw#58

Merged

cmsbuild added code-checks-rejected and removed code-checks-pending labels Oct 13, 2021

cmsbuild added code-checks-pending and removed code-checks-rejected labels Oct 13, 2021

cmsbuild added code-checks-approved and removed code-checks-pending labels Oct 13, 2021

cmsbuild added requires-external tests-started and removed tests-pending labels Oct 13, 2021

cmsbuild added tests-approved and removed tests-started labels Oct 13, 2021

cmsbuild added fully-signed reconstruction-approved and removed reconstruction-pending pending-signatures labels Oct 15, 2021

mmasciov mentioned this pull request Oct 17, 2021

Enable mkFit by default in 4+2 tracking iterations (initialStepPreSpl… #35660

Merged

cmsbuild added orp-approved and removed orp-pending labels Oct 19, 2021

cmsbuild merged commit fd8dd86 into cms-sw:master Oct 19, 2021

cmsbuild mentioned this pull request Oct 19, 2021

Migrate to esConsumes muon alignment #35719

Merged

mmasciov mentioned this pull request Nov 3, 2021

Update mkFit for CMSSW_12_2_0 #35974

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update mkFit for 12_1_0_pre5 #35652

Update mkFit for 12_1_0_pre5 #35652

mmasciov commented Oct 13, 2021 •

edited

Loading

cmsbuild commented Oct 13, 2021

slava77 commented Oct 13, 2021

makortel commented Oct 13, 2021

slava77 commented Oct 13, 2021

makortel commented Oct 13, 2021

makortel commented Oct 13, 2021 •

edited

Loading

cmsbuild commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

makortel commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

makortel commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

jpata commented Oct 15, 2021

cmsbuild commented Oct 15, 2021

perrotta commented Oct 19, 2021

Update mkFit for 12_1_0_pre5 #35652

Update mkFit for 12_1_0_pre5 #35652

Conversation

mmasciov commented Oct 13, 2021 • edited Loading

PR description:

PR validation:

cmsbuild commented Oct 13, 2021

slava77 commented Oct 13, 2021

makortel commented Oct 13, 2021

slava77 commented Oct 13, 2021

makortel commented Oct 13, 2021

makortel commented Oct 13, 2021 • edited Loading

cmsbuild commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

makortel commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

makortel commented Oct 13, 2021

cmsbuild commented Oct 13, 2021

Comparison Summary

jpata commented Oct 15, 2021

cmsbuild commented Oct 15, 2021

perrotta commented Oct 19, 2021

mmasciov commented Oct 13, 2021 •

edited

Loading

makortel commented Oct 13, 2021 •

edited

Loading