Adding UFO calibration, new CDI, fixing Xbb cut
What does this MR do?:
- Adding a user-configured jet calibration tool for UFO jets when running on Data to avoid SUSYTools crash.
- updating to new CDI and jet uncertainties recommendation
- Adding an OR requirement for track jet based b-tagging and the Xbb_25 80% WP. I believe that in our current ntuples (v05) the preselection is not quite correct for doing Xbb-tagger studies because the preselection applied here is always selecting events which have 2 b-tagged VR trk jets associated to the large R jet. So applying the Xbb tagger also always applies the track jet b-tagging.
Problems
I tested 1. on a data18 sample file from the p4321 EXOT27 and it worked fine. When I tried to run on a file from Znunu_PTV500_1000 (364222) p4320 I got the following error:
Click to see error
xAOD::TAuxStore::readFrom WARNING Static branch (EventInfoAux.) with split level 0 discovered
xAOD::TAuxStore::readFrom WARNING The reading of complex variables from it may/will fail!
IncidentSvc ERROR Standard std::exception is caught handling incident0x7ffe295f0d98
IncidentSvc ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
IncidentSvc ERROR Standard std::exception is caught handling incident0x7ffe295f0d98
IncidentSvc ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
IncidentSvc ERROR Standard std::exception is caught handling incident0x7ffe295f0d98
IncidentSvc ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
IncidentSvc ERROR Standard std::exception is caught handling incident0x7ffe295f0d98
IncidentSvc ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
IncidentSvc ERROR Standard std::exception is caught handling incident0x7ffe295f0d98
IncidentSvc WARNING ERROR message limit (10) reached for IncidentSvc. Suppressing further output.
xAOD::TAuxStore::readFrom WARNING Static branch (PrimaryVerticesAux.) with split level 0 discovered
xAOD::TAuxStore::readFrom WARNING The reading of complex variables from it may/will fail!
ToolSvc.TrigDecisionTool INFO updating config with SMK: 2216 and L1PSK: 76 and HLTPSK: 260
DecorateLRTruthLabels FATAL Standard std::exception is caught
DecorateLRTruthLabels ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
AthAlgSeq FATAL Standard std::exception is caught
AthAlgSeq ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
AthMasterSeq FATAL Standard std::exception is caught
AthMasterSeq ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297).
Traceback (most recent call last):
File "/cvmfs/atlas.cern.ch/repo/sw/software/21.2/AthAnalysis/21.2.150/InstallArea/x86_64-centos7-gcc8-opt/jobOptions/AthenaCommon/runbatch.py", line 20, in <module>
theApp.run() # runs until theApp.EvtMax events reached
File "/cvmfs/atlas.cern.ch/repo/sw/software/21.2/AthAnalysis/21.2.150/InstallArea/x86_64-centos7-gcc8-opt/python/AthenaCommon/AppMgr.py", line 663, in run
sc = self.getHandle()._evtpro.executeRun( nEvt )
Exception: StatusCode IEventProcessor::executeRun(int maxevt) =>
SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::eventTypeBitmask' (1297). (C++ exception of type SG::ExcBadAuxVar)
XAMPPAlgorithm INFO Finalizing XAMPPAlgorithm...
MetaDataMC::SaveMetaDa... INFO Write metadata in file with DSID: 364222, ProcessID : 0, process name: Nominal_Inclusive SumW: 14279.906750, SumW2: 38391.967182, TotalEvents: 20000, ProcessedEvents: 0
TreeBase::FinalizeTree() INFO Successfully written MonoH_Nominal containing 0 entries.
This can be fixed by changing AssembleIO() to AssembleIO(1)
, i.e. changing access mechanism from kBranchAccess to kClassAccess.
However, even with this fix the code breaks after a couple of thousand events have been processed with the following error:
Click to see error
AthenaEventLoopMgr INFO ===>>> done processing event #0, run #0 1 events processed so far <<<===
Entry 1000 / 20000 (5%) in file 1 / 1 @ 00:00:30. Physics Event Rate: 33.1 Hz, Computing Event Rate: 33.1 Hz, E.T.A.: 00:09:33 (updating screen each 1000 events)
Entry 2000 / 20000 (10%) in file 1 / 1 @ 00:00:58. Physics Event Rate: 34.2 Hz, Computing Event Rate: 34.2 Hz, E.T.A.: 00:08:45 (updating screen each 1000 events)
Entry 3000 / 20000 (15%) in file 1 / 1 @ 00:01:26. Physics Event Rate: 34.7 Hz, Computing Event Rate: 34.7 Hz, E.T.A.: 00:08:10 (updating screen each 1000 events)
Entry 4000 / 20000 (20%) in file 1 / 1 @ 00:01:54. Physics Event Rate: 34.8 Hz, Computing Event Rate: 34.8 Hz, E.T.A.: 00:07:39 (updating screen each 1000 events)
Entry 5000 / 20000 (25%) in file 1 / 1 @ 00:02:22. Physics Event Rate: 35 Hz, Computing Event Rate: 35 Hz, E.T.A.: 00:07:08 (updating screen each 1000 events)
Entry 6000 / 20000 (30%) in file 1 / 1 @ 00:02:51. Physics Event Rate: 35 Hz, Computing Event Rate: 35 Hz, E.T.A.: 00:06:39 (updating screen each 1000 events)
Entry 7000 / 20000 (35%) in file 1 / 1 @ 00:03:19. Physics Event Rate: 35 Hz, Computing Event Rate: 35 Hz, E.T.A.: 00:06:11 (updating screen each 1000 events)
XAMPPAlgorithm FATAL Standard std::exception is caught
XAMPPAlgorithm ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::pdgId' (189).
AthAlgSeq FATAL Standard std::exception is caught
AthAlgSeq ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::pdgId' (189).
AthMasterSeq FATAL Standard std::exception is caught
AthMasterSeq ERROR SG::ExcBadAuxVar: Attempt to retrieve nonexistent aux data item `::pdgId' (189).
Traceback (most recent call last):
File "/cvmfs/atlas.cern.ch/repo/sw/software/21.2/AthAnalysis/21.2.150/InstallArea/x86_64-centos7-gcc8-opt/jobOptions/AthenaCommon/runbatch.py", line 20, in <module>
theApp.run() # runs until theApp.EvtMax events reached
File "/cvmfs/atlas.cern.ch/repo/sw/software/21.2/AthAnalysis/21.2.150/InstallArea/x86_64-centos7-gcc8-opt/python/AthenaCommon/AppMgr.py", line 663, in run
sc = self.getHandle()._evtpro.executeRun( nEvt )
The behaviour is the same for a Wenu_500_1000 (364182) file for which I have tested this. Could it be that there is something wrong with the EXOT27 derivations produced with p4320? Just to be clear I haven't seen these errors with the following EXOT27 productions: p3990 (364222 MonoH), p3991 (signal MonoH), p4050 (MonoH data), p4321 (data with UFO).
How to proceed?
There are 3 things which I would like to check the impact of and these are:
- UFO vs LCTopo
- 201810 vs 201903 vs Xbb b-tagging
- emu CR
The point 1. seems difficult because in the case that the problem with p4320 is not a problem of XAMPP we would have to wait for new derivations.
Point 3. I would like to address in our meeting tomorrow. This is also the reason I put this MR as WIP for the moment because I want to make some changes here before submitting a new ntuple production.
For this reason, I would propose that I produce a new set of ntuples with p3990 & p4062 (this is what we used for v05) but using 201903 b-tagging and the correct preselection for Xbb b-tagging which I could then compare to the Mono-H selection of the v05 ntuples addressing 2. and 3..