Removing "Mult" hypotools as discussed in ATR-21779

As tidying up triEgammaHypo which no longer needs "Mult" hypotools; although following tests have reference missmatch

INFO     --------------------------------------------------------------------------------------------------------------------------------------------
INFO     | test_trigAna_RDOtoRDOTrig_mt1_build.py | Reco_tf: 0, CountRefComp: 1, CheckLog: 0, MessageCount: 3, RootComp: 48, ZeroCounts: 0, exit: 1 |
INFO     | test_trigP1_v1Dev_build.py             | athenaHLT: 0, CheckLog: 0, MessageCount: 2, RootComp: 76, ZeroCounts: 0, exit: 0                |
INFO     | test_trig_data_newJO_build.py          | GeneratePickle: 0, ExecSummary: 0, exit: 0                                                      |
INFO     | test_trig_data_v1Dev_build.py          | athena: 0, CheckLog: 0, MessageCount: 0, RootComp: 77, ZeroCounts: 0, CountRefComp: 1, exit: 1  |
INFO     --------------------------------------------------------------------------------------------------------------------------------------------
INFO     2 tests succeeded out of 4 executed
ERROR    ==================================================
ERROR    The following 2 tests failed:
ERROR        test_trigAna_RDOtoRDOTrig_mt1_build.py
ERROR        test_trig_data_v1Dev_build.py
ERROR    ==================================================

the details of mismatch can be found here /afs/cern.ch/work/d/dbakshig/public/ElectronHypoSan/run/runTrigART/results

An example of reference mismatch runTrigART/results/runTrigART/test_trigAna_RDOtoRDOTrig_mt1_build/CountRefComp.log

INFO HLT_2e17_lhvloose_L12EM15VH #2280638160
-TrigSignatureMoniMT                                INFO -- #2280638160 Events         2          2          0          0          0          0          -          -          0
-TrigSignatureMoniMT                                INFO -- #2280638160 Features                             0          0          0          0          -          -
+TrigSignatureMoniMT                                INFO -- #2280638160 Events         2          2          2          0          0          0          -          -          0
+TrigSignatureMoniMT                                INFO -- #2280638160 Features                             2          0          0          0          -          -

Is there a way to figure out that my changes are not adding anything unusual. Adding @tamartin , @fernando , @cjmeyer Thanks, Debo.

The change in the log could be benign.

E.g. before we have two events which seed the chain. But if we only find 1 fast cluster satisfying pT > 17 in each of these two event, then as 1 < 2 your HypoAlg used to reject this. So we saw zero events, zero 'features' at Step1.

Now... your HypoAlg would pass the single fast-cluster in each of the two Events, and the ComboHypoAlg would be the component which handles the "1 < 2 therefore reject" logic.

I think we are overdue an audit of how the TrigSignatureMoniMT interacts with both the HypoAlg and ComboHypoAlg in a given step, as we've seen similar oddities in other slices.

@dbakshig - it would be good to double-check this hypothesis with DEBUG level output.

Thanks Tim! @dbakshig to check debug, try using the ZEE DRAW from data as in TrigEgammaSoftwareRun3 twiki:

/eos/atlas/user/f/fernando/./data17_13TeV/data17_13TeV/data17_13TeV.00339070 .physics_Main.merge.DRAW_EGZ.f887_m1831._0049.1

Cheers!

Fer

Hi @tamartin ,

I printed Hypos in DEBUG mode also the chain count differences are in sync as you mentioned but again I couldn't get any exact message for how ComboHypo rejecting the events; e.g.

 TrigSignatureMoniMT                                 INFO HLT_2e17_etcut_L12EM15VH #3136730292
-TrigSignatureMoniMT                                 INFO -- #3136730292 Events         20         20         0          0          0          -          -          -          0
-TrigSignatureMoniMT                                 INFO -- #3136730292 Features                             0          0          0          -          -          -
+TrigSignatureMoniMT                                 INFO -- #3136730292 Events         20         20         4          0          0          -          -          -          0
+TrigSignatureMoniMT                                 INFO -- #3136730292 Features                             4          0          0          -          -          -

You can see the log file for this here /afs/cern.ch/work/d/dbakshig/public/ElectronHypoSan/run/runTrigART/results/runTrigART/test_trig_data_v1Dev_build/CountRefComp.log is there a place I get more insight?

Thanks, Debo.

So what do we see... in event two it looks like this HypoTool sees one electron. And passes it.

ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG AcceptAll property not set: applying selection
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG ; RoI ID = 0: Eta = -0.1, Phi = -0.392699
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: eta=-0.18522 roi eta=-0.1 DeltaEta=-0.0852195 cut: < 'dETACLUSTERthr':9999.00
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG : phi=-0.42186 roi phi=-0.392699 DeltaPhi=0.0291609 cut: < 'dPHICLUSTERthr':9999.00
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG eta bin used for cuts 0
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: Rcore=-0.580376 cut: >-9999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG  cut: >-9999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: ET_em=462.705 cut: >0
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG ET_em<90000: use low ET_had cut: <9999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: ET_had=-0.159349 cut: <9999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: F1=0.243397 cut: >0
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: Weta2=0.019742 cut: <99999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: Wstot=11.5766 cut: <99999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG TrigEMCluster: F3=0.240705 cut: <99999
ElectronEgammaFastCaloHypo.HLT_2e3_et...   2   0   DEBUG pass = 1
...
ElectronEgammaFastCaloHypo                 2   0   DEBUG Exiting with 1 Decision objects
ElectronEgammaFastCaloHypo                 2   0   DEBUG Number of positive decisions for Decision object #0: 4
ElectronEgammaFastCaloHypo                 2   0   DEBUG  --- Passes: leg000_HLT_e3_etcut1step_mu6fast_L1EM8I_MU10 ID#135406112
ElectronEgammaFastCaloHypo                 2   0   DEBUG  --- Passes: HLT_e3_etcut_L1EM3 ID#683953566
ElectronEgammaFastCaloHypo                 2   0   DEBUG  --- Passes: HLT_2e3_etcut_L12EM3 ID#2613484113
ElectronEgammaFastCaloHypo                 2   0   DEBUG  --- Passes: leg000_HLT_e3_etcut1step_mu26_L1EM8I_MU10 ID#3190265273

This is fine - the ComboHypo should reject this single electron passing HLT_2e3_etcut_L12EM3

And we see the ComboHypo here

                Step1_2FastCalo_electron [Seq]  [Sequential]  [Prompt]
                  FStep1_2FastCalo_electron [Alg]  [n= 0]
                  Step1_2FastCalo_electron_view [Seq]  [Sequential]  [Prompt]
                    Step1_2FastCalo_electron_reco [Seq]  [Concurrent]  [OR]
                      fastCaloSequence [Seq]  [Sequential]  [Prompt]
                        IMfastCalo [Alg]  [n= 0]
                        fastCaloInViewSequence [Seq]  [Sequential]  [Prompt]
                          fastCaloVDV [Alg]  [n= 0]
                          FastCaloL2EgammaAlg [Alg]  [n= 0]
                    ElectronEgammaFastCaloHypo [Alg]  [n= 0]
                    ComboHypo_Step1_2FastCalo_electron [Alg]  [n= 0]

But we don't have any DEBUG level output from this alg.

Did you only turn on DEBUG for some algs? Maybe could run with DEBUG turned on for everything, but only egamma slice (to keep it a bit more manageable log size!)

added 2902 commits

• 38c8a23b...e739c48f - 2897 commits from branch atlas:master
• a2a185d8 - Merge remote-tracking branch 'upstream/master' into removingMultHypoTool
• 9a098cc6 - Adding DEBUG outputLevel
• 99d4608d - Merge remote-tracking branch 'upstream/master' into removingMultHypoTool
• c821433a - Merge remote-tracking branch 'upstream/master' into removingMultHypoTool
• f3fe1762 - Switching some DEBUG messages

Compare with previous version

mentioned in merge request !36500 (closed)

added 714 commits

• f3fe1762...d2a55f48 - 711 commits from branch atlas:master
• 63398b55 - disabling DEBUG messages
• ab35a128 - Merge remote-tracking branch 'upstream/master' into removingMultHypoTool
• d0e80491 - updating references

Compare with previous version

Hi @tamartin ,

Thanks for your comments.

Debo.

unmarked as a Work In Progress

changed the description

WARNING: big files (>100K) are found in the changeset

172K in file Trigger/TrigValidation/TrigAnalysisTest/share/ref_RDOtoRDOTrig_v1Dev_build.ref

176K in file Trigger/TrigValidation/TriggerTest/share/ref_data_v1Dev_build.ref

This merge request affects 4 packages:

• Trigger/TrigHypothesis/TrigEgammaHypo
• Trigger/TrigValidation/TrigAnalysisTest
• Trigger/TrigValidation/TriggerTest
• Trigger/TriggerCommon/TriggerMenuMT

Adding @okumura ,@bernius ,@jpanduro as watchers

added Egamma Trigger TriggerMenu master review-pending-level-1 labels

CI Result SUCCESS (hash d0e80491)

	Athena	AthSimulation	AthGeneration	AnalysisBase
externals
cmake
make
required tests
optional tests

Full details available on this CI monitor view
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
AthGeneration: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 20886]

added review-approved label and removed review-pending-level-1 label

Hi @dbakshing, I have no comments on the code, It looks okay and since the CI ran successfully I approve.

Cheers, Santiago (L1)

Dear shifter can it be merged now?

Thanks

resolved all threads

merged

mentioned in commit cb993dca

changed the description

added sweep:ignore label