Skip to content

Backport of Muon L1 prefire weights #33759

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cmsbuild merged 20 commits into from
Jun 12, 2021
Merged

Backport of Muon L1 prefire weights #33759

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
20 commits
Select commit Hold shift + click to select a range
f475bcb
backport muon prefiring weights to 106X
JanFSchulte May 17, 2021
3db916c
undo unwanted changes
JanFSchulte May 17, 2021
4f60410
make even with PR to master
JanFSchulte May 27, 2021
50f940b
fix configs for miniAOD
JanFSchulte May 27, 2021
2fac91c
code format and checks
JanFSchulte May 27, 2021
447aafc
protect prefiring weights for nano V8
JanFSchulte May 28, 2021
d752ac6
fix typo
JanFSchulte May 30, 2021
97c3c7c
fix another typo
JanFSchulte May 31, 2021
67afdb9
update to 2016 UL prefiring maps for ECAL
JanFSchulte Jun 1, 2021
de785c4
fix swapped stat and syst uncertainties
JanFSchulte Jun 2, 2021
34fdaef
add prefiring weights to Nano DQM
JanFSchulte Jun 2, 2021
c7471ec
revert changes to miniAOD config
JanFSchulte Jun 2, 2021
fd55306
set JetMaxMuonFraction to -1 for miniAOD
JanFSchulte Jun 2, 2021
3c05ba7
make prefiring weights in nano doubles again
JanFSchulte Jun 3, 2021
16ca043
implement suggestions from Laurent
JanFSchulte Jun 4, 2021
9a4afae
bug fix and merge jet and photon weights
JanFSchulte Jun 4, 2021
5a0e55d
bug fix to nano config
JanFSchulte Jun 6, 2021
6267072
fix DQM to not have weights of 1 in the overflow
JanFSchulte Jun 8, 2021
c848b76
remove empty plots from DQM and protect 2016 weights in V8
JanFSchulte Jun 8, 2021
991fa3b
fix weight labels in nano
JanFSchulte Jun 9, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions PhysicsTools/NanoAOD/python/nanoDQM_cff.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -142,6 +142,14 @@
_fatjet_plots_106Xv1.append(plot)
(run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(nanoDQM.vplots.FatJet, plots = _fatjet_plots_106Xv1 )

_prefire_plots_nom = copy.deepcopy(nanoDQM.vplots.L1PreFiringWeight.plots)
_prefire_plots_106Xv1 = cms.VPSet()
for plot in _prefire_plots_nom:
if (not "Muon" in plot.name.value() and not "ECAL" in plot.name.value()):
_prefire_plots_106Xv1.append(plot)
(run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(nanoDQM.vplots.L1PreFiringWeight, plots = _prefire_plots_106Xv1 )



## MC
nanoDQMMC = nanoDQM.clone()
Expand Down
11 changes: 11 additions & 0 deletions PhysicsTools/NanoAOD/python/nanoDQM_cfi.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -809,5 +809,16 @@
Plot1D('rawMVAoldDMdR032017v2', 'rawMVAoldDMdR032017v2', 20, -1, 1, 'byIsolationMVArun2017v2DBoldDMdR0p3wLT raw output discriminator (2017v2)'),
)
),
L1PreFiringWeight = cms.PSet(
sels = cms.PSet(),
plots = cms.VPSet(
Plot1D('Nom', 'Nom', 21, 0.8, 1.01, 'L1 prefiring weight nominal'),
Plot1D('Up', 'Up', 21, 0.8, 1.01, 'L1 prefiring weight uncertainy up'),
Plot1D('Dn', 'Dn', 21, 0.8, 1.01, 'L1 prefiring weight uncertainty down'),
Plot1D('ECAL_Nom', 'ECAL_Nom', 21, 0.8, 1.01, 'L1 prefiring weight for ECAL objects nominal'),
Plot1D('Muon_Nom', 'Muon_Nom', 21, 0.8, 1.01, 'L1 prefiring weight for muons nominal'),
)

),
)
)
5 changes: 5 additions & 0 deletions PhysicsTools/NanoAOD/python/nano_eras_cff.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,10 +4,15 @@
from Configuration.Eras.Modifier_run2_jme_2016_cff import run2_jme_2016
from Configuration.Eras.Modifier_run2_jme_2017_cff import run2_jme_2017
from Configuration.Eras.Modifier_run2_muon_2016_cff import run2_muon_2016
from Configuration.Eras.Modifier_run2_muon_2018_cff import run2_muon_2018

from Configuration.Eras.Modifier_run2_tau_ul_2016_cff import run2_tau_ul_2016
from Configuration.Eras.Modifier_run2_tau_ul_2018_cff import run2_tau_ul_2018

from Configuration.Eras.Modifier_run2_HLTconditions_2016_cff import run2_HLTconditions_2016
from Configuration.Eras.Modifier_run2_HLTconditions_2017_cff import run2_HLTconditions_2017
from Configuration.Eras.Modifier_run2_HLTconditions_2018_cff import run2_HLTconditions_2018

from Configuration.Eras.Modifier_run2_miniAOD_80XLegacy_cff import run2_miniAOD_80XLegacy
from Configuration.Eras.Modifier_run2_nanoAOD_92X_cff import run2_nanoAOD_92X
from Configuration.Eras.Modifier_run2_nanoAOD_94XMiniAODv1_cff import run2_nanoAOD_94XMiniAODv1
Expand Down
47 changes: 37 additions & 10 deletions PhysicsTools/NanoAOD/python/triggerObjects_cff.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,8 +1,6 @@
import FWCore.ParameterSet.Config as cms
from Configuration.Eras.Modifier_run2_HLTconditions_2016_cff import run2_HLTconditions_2016
from Configuration.Eras.Modifier_run2_HLTconditions_2017_cff import run2_HLTconditions_2017
from Configuration.Eras.Modifier_run2_miniAOD_80XLegacy_cff import run2_miniAOD_80XLegacy
from PhysicsTools.NanoAOD.common_cff import ExtVar
from PhysicsTools.NanoAOD.nano_eras_cff import *
from PhysicsTools.NanoAOD.common_cff import *
import copy

unpackedPatTrigger = cms.EDProducer("PATTriggerObjectStandAloneUnpacker",
Expand Down Expand Up @@ -218,19 +216,48 @@
selections = selections2016
)

from PhysicsTools.PatUtils.L1ECALPrefiringWeightProducer_cff import prefiringweight
run2_HLTconditions_2016.toModify(prefiringweight, DataEra = cms.string("2016BtoH"))
from PhysicsTools.PatUtils.L1PrefiringWeightProducer_cff import prefiringweight
#Next lines are for UL2016 maps
(run2_muon_2016 & tracker_apv_vfp30_2016).toModify( prefiringweight, DataEraECAL = cms.string("UL2016preVFP"), DataEraMuon = cms.string("2016preVFP"))
(run2_muon_2016 & ~tracker_apv_vfp30_2016).toModify( prefiringweight, DataEraECAL = cms.string("UL2016postVFP"), DataEraMuon = cms.string("2016postVFP"))
#Next line is for UL2017 maps
run2_jme_2017.toModify( prefiringweight, DataEraECAL = cms.string("UL2017BtoF"), DataEraMuon = cms.string("20172018"))
#Next line is for UL2018 maps
run2_muon_2018.toModify( prefiringweight, DataEraECAL = cms.string("None"), DataEraMuon = cms.string("20172018"))

#For pre-UL 2017 reprocessing, one should use the original maps and no muon jet protection
for modifier in run2_nanoAOD_94XMiniAODv1, run2_nanoAOD_94XMiniAODv2:
modifier.toModify( prefiringweight, DataEraECAL = cms.string("2017BtoF"), DataEraMuon = cms.string("20172018"))
modifier.toModify( prefiringweight, JetMaxMuonFraction = cms.double(-1.) )
#For pre-UL 2016 reprocessing, same thing
run2_nanoAOD_94X2016.toModify( prefiringweight, DataEraECAL = cms.string("2016BtoH"), DataEraMuon = cms.string("2016") )
run2_nanoAOD_94X2016.toModify( prefiringweight, JetMaxMuonFraction = cms.double(-1.) )

l1PreFiringEventWeightTable = cms.EDProducer("GlobalVariablesTableProducer",
name = cms.string("L1PreFiringWeight"),
variables = cms.PSet(
L1PreFiringWeight_Nom = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProb"), "double", doc = "L1 pre-firing event correction weight (1-probability)", precision=8),
L1PreFiringWeight_Up = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbUp"), "double", doc = "L1 pre-firing event correction weight (1-probability), up var.", precision=8),
L1PreFiringWeight_Dn = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbDown"), "double", doc = "L1 pre-firing event correction weight (1-probability), down var.", precision=8),
Nom = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProb"), "double", doc = "L1 pre-firing event correction weight (1-probability)", precision=8),
Up = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbUp"), "double", doc = "L1 pre-firing event correction weight (1-probability), up var.", precision=8),
Dn = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbDown"), "double", doc = "L1 pre-firing event correction weight (1-probability), down var.", precision=8),
Muon_Nom = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbMuon"), "double", doc = "Muon L1 pre-firing event correction weight (1-probability)", precision=8),
Muon_SystUp = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbMuonSystUp"), "double", doc = "Muon L1 pre-firing event correction weight (1-probability), up var. syst.", precision=8),
Muon_SystDn = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbMuonSystDown"), "double", doc = "Muon L1 pre-firing event correction weight (1-probability), down var. syst.", precision=8),
Muon_StatUp = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbMuonStatUp"), "double", doc = "Muon L1 pre-firing event correction weight (1-probability), up var. stat.", precision=8),
Muon_StatDn = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbMuonStatDown"), "double", doc = "Muon L1 pre-firing event correction weight (1-probability), down var. stat.", precision=8),
ECAL_Nom = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbECAL"), "double", doc = "ECAL L1 pre-firing event correction weight (1-probability)", precision=8),
ECAL_Up = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbECALUp"), "double", doc = "ECAL L1 pre-firing event correction weight (1-probability), up var.", precision=8),
ECAL_Dn = ExtVar(cms.InputTag("prefiringweight:nonPrefiringProbECALDown"), "double", doc = "ECAL L1 pre-firing event correction weight (1-probability), down var.", precision=8),
)
)

triggerObjectTables = cms.Sequence( unpackedPatTrigger + triggerObjectTable )

#protect nano V8 production from changes to prefiring weights introduce for V9
(run2_jme_2016 & run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(prefiringweight,DataEraECAL = cms.string("2016BtoH"))
(run2_jme_2017 & run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(prefiringweight,DataEraECAL = cms.string("2017BtoF"))
(run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(prefiringweight,DoMuons = cms.bool(False))
(run2_nanoAOD_106Xv1 & ~run2_nanoAOD_devel).toModify(l1PreFiringEventWeightTable.variables , Muon_Nom = None, Muon_SystUp = None, Muon_SystDn = None, Muon_StatUp = None, Muon_StatDn = None, ECAL_Nom = None, ECAL_Up = None, ECAL_Dn = None)

Copy link
Contributor Author

@JanFSchulte JanFSchulte Jun 8, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@mariadalfonso Here we remove the new prefiring weights from the nano event content for V8, so having empty histograms for them in DQM is expected. What I am a puzzled about is why the prefiring weight is 1, it should still reflect the ECAL prefiring. What workflow did you use for that test?

Copy link
Contributor Author

@JanFSchulte JanFSchulte Jun 8, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mhhh, now that we have the UL maps for ECAL, this might need a further protection for that year. @lathomas

_triggerObjectTables_withL1PreFiring = triggerObjectTables.copy()
_triggerObjectTables_withL1PreFiring.replace(triggerObjectTable, prefiringweight + l1PreFiringEventWeightTable + triggerObjectTable)
(run2_HLTconditions_2016 | run2_HLTconditions_2017).toReplaceWith(triggerObjectTables, _triggerObjectTables_withL1PreFiring)
(run2_HLTconditions_2016 | run2_HLTconditions_2017 | run2_HLTconditions_2018).toReplaceWith(triggerObjectTables, _triggerObjectTables_withL1PreFiring)
8 changes: 4 additions & 4 deletions PhysicsTools/PatAlgos/python/slimming/miniAOD_tools.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -554,11 +554,11 @@ def miniAOD_customizeCommon(process):

# L1 pre-firing weights for 2016 and 2017
from Configuration.Eras.Modifier_run2_L1prefiring_cff import run2_L1prefiring
from Configuration.Eras.Modifier_stage1L1Trigger_cff import stage1L1Trigger
from Configuration.Eras.Modifier_stage2L1Trigger_cff import stage2L1Trigger
from Configuration.Eras.Modifier_stage2L1Trigger_2017_cff import stage2L1Trigger_2017
process.load("PhysicsTools.PatUtils.L1ECALPrefiringWeightProducer_cff")
stage1L1Trigger.toModify(process.prefiringweight, DataEra = "2016BtoH")
stage2L1Trigger_2017.toModify(process.prefiringweight, DataEra = "2017BtoF")
process.load("PhysicsTools.PatUtils.L1PrefiringWeightProducer_cff")
stage2L1Trigger.toModify(process.prefiringweight, DataEraECAL = "2017BtoF", JetMaxMuonFraction = -1, DoMuons = cms.bool(False) )
Copy link
Contributor

@perrotta perrotta Jun 10, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the era correct here ("2017BtoF")?

Copy link
Contributor

@lathomas lathomas Jun 10, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

if we want to reproduce the configuration previous to this PR to comply with the rule forbidding changes to RECO, yes. (these 2017 maps were also mistakenly picked previously for 2016 ...)

Copy link
Contributor

@perrotta perrotta Jun 10, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

if we want to reproduce the configuration previous to this PR to comply with the rule forbidding changes to RECO, yes. (these 2017 maps were also mistakenly picked previously for 2016 ...)

Ok, thank you!

stage2L1Trigger_2017.toModify(process.prefiringweight, DataEraECAL = "2017BtoF", JetMaxMuonFraction = -1, DoMuons = cms.bool(False))
Comment on lines -559 to +561
Copy link
Contributor

@slava77 slava77 Jun 8, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is this suppose to lead to any changes in the production miniAOD (especially UL)?
I see no differences in cms-bot reco comparisons, but we do not have UL miniAOD tests for 2016 and 2017 in these comparisons.

Copy link
Contributor Author

@JanFSchulte JanFSchulte Jun 8, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No, this is specifically set to prevent any changes to miniAOD by deactivating the new features and using the pre-UL ECAL maps.

Copy link
Contributor

@mariadalfonso mariadalfonso Jun 10, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do we need to have something set for the stage2L1Trigger_2018?
this is what is done in master
https://github.com/cms-sw/cmssw/pull/33758/files#diff-c1409af99a0dfee10534eb0c4e985d3b4dd669b5c81cd302efdf760ee0ce5c41R550

Copy link
Contributor Author

@JanFSchulte JanFSchulte Jun 10, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the same rule of not changing the miniAOD output applies here, not thing was set for 2018 before, so we set nothing now.

run2_L1prefiring.toModify(task, func=lambda t: t.add(process.prefiringweight))

def miniAOD_customizeMC(process):
Expand Down
226 changes: 0 additions & 226 deletions PhysicsTools/PatUtils/plugins/L1ECALPrefiringWeightProducer.cc
View file
Open in desktop

This file was deleted.

Loading