fix: Propagate initial vertex time variance for AMVF w/o time

[andiwand]

Logs like

21:16:17    VertexPerfor   WARNING   Nonpositive variance after vertex fit: Var(T) = 0 <= 0.

are gone after this change.

To me it seems more sensical to propagate the initial time variance instead of hard setting it to 0 which causes the cov matrix to be invalid.

[andiwand]

@felix-russo do you have an opinion on this?

[pbutti]

@andiwand Alternatively one could make sure that a very large positive value to the initial for sigma_tt is present in the initial cov matrix (or set it to a very large value) and check/set that the other terms in the last row/column of the covariance matrix to 0 and just call the KF update on it.
Removing the lines where the time row/column are set to 0 makes sense to me.
In the case of no time information, the last row/column should not provide any weight and setting it as above should achieve that.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (86e62ab) 48.84% compared to head (d8af9fb) 48.84%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2936   +/-   ##
=======================================
  Coverage   48.84%   48.84%           
=======================================
  Files         495      495           
  Lines       28911    28909    -2     
  Branches    13734    13732    -2     
=======================================
  Hits        14122    14122           
  Misses       4890     4890           
+ Partials     9899     9897    -2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[github-actions]

📊: Physics performance monitoring for d8af9fb

Full contents

physmon summary

• ✅ CKF truth_smeared
• ✅ IVF truth_smeared
• ✅ AMVF truth_smeared
• ✅ Track Summary CKF truth_smeared
• ✅ Seeding truth_estimated
• ✅ CKF truth_estimated
• ✅ IVF truth_estimated
• ✅ AMVF truth_estimated
• ✅ Track Summary CKF truth_estimated
• ✅ Seeding seeded
• ✅ CKF seeded
• ✅ IVF seeded
• ✅ AMVF seeded
• ✅ AMVF (+grid seeder) seeded
• ✅ Track Summary CKF seeded
• ✅ Seeding orthogonal
• ✅ CKF orthogonal
• ✅ IVF orthogonal
• ✅ AMVF orthogonal
• ✅ Track Summary CKF orthogonal
• ✅ Ambisolver seeded
• ✅ Ambisolver orthogonal
• ✅ Seeding ttbar
• ✅ CKF ttbar
• ✅ Ambisolver
• ✅ Track Summary CKF ttbar
• ✅ AMVF ttbar
• ✅ AMVF (+grid seeder) ttbar
• ✅ Truth tracking (GSF)
• ✅ Truth tracking
• ✅ Particles fatras
• ✅ Particles geant4

[andiwand]

@andiwand Alternatively one could make sure that a very large positive value to the initial for sigma_tt is present in the initial cov matrix (or set it to a very large value) and check/set that the other terms in the last row/column of the covariance matrix to 0 and just call the KF update on it. Removing the lines where the time row/column are set to 0 makes sense to me. In the case of no time information, the last row/column should not provide any weight and setting it as above should achieve that.

@pbutti We are actually forwarding the initial covariance into the vertex fit which, like you described, has a big diagonal covariance. To me it seems good enough to use the initial cov. What do you think?

[felix-russo]

@felix-russo do you have an opinion on this?

Looks very good to me!

[andiwand]

I will go ahead and update the reference here

[acts-project-service]

🔴 Athena integration test results [b50dd08]

🔴 Some tests have failed!

Please investigate the pipeline!

status	job	report
🟢	run_unit_tests
🟢	test_ActsEFTrackFit
🟢	test_ActsBenchmarkWithSpot
🟢	test_ActsConversionWorkflow
🟢	test_ActsWorkflow
🔴	test_ActsValidateAmbiguityResolution
🟢	test_ActsValidateResolvedTracks
🟢	test_ActsValidateTracks
🟢	test_ActsValidateActsCoreSpacePoints
🟢	test_ActsValidateActsSpacePoints
🔴	test_ActsValidateSeeds
🟢	test_ActsValidateOrthogonalSeeds
🟢	test_ActsValidateClusters
🟢	test_ActsPersistifyEDM
🔴	test_ActsGSFRefitting
🟢	test_ActsKfRefitting
🟢	test_ActsExtrapolationAlgTest
🟢	test_ActsITkTest
🟢	run_workflow_tests_run4_mc
🟢	run_workflow_tests_run2_mc
🟢	run_workflow_tests_run2_data
🔴	run_workflow_tests_run3_mc
🟢	run_workflow_tests_run3_data
🟢	run_art_test: test_data18_13TeV_1000evt
🟢	run_art_test: test_ttbarPU40_reco

[paulgessinger]

Athena failures unrelated.