def connect(self):
'''Retrieves the relevant information
(e.g. files location) for each component.
To avoid multiple connections to the database
is production == True, it checks for a cached
pickle file containing all the info.
FIXME! RIC: this should be done by default,
but I make the use of the cached pickle explicit
because the name of the parent dataset, and
therefore the number of events and the
computing efficiency, is not
saved in the pickle file, and it can only be retrieved
through a query to the database. This is
necessary at the analysis level, but not at
the production stage, where the only bit of info
that's really needed is the location of the files.
Should revisit the way the pickle file is saved
so that ALL the relevant info is stored there.
'''
if self.production:
self.connect_by_pck_()
else:
self.connect_by_db_()
self.pruneSampleList_()
for sample in self.MC_list:
sample.splitFactor = splitFactor(sample, self.splitFactor)
def listify_(self):
''' '''
self.MC_list = [v for k, v in self.mc_dict.items()]
for sam in self.MC_list:
sam.splitFactor = splitFactor(sam, self.splitFactor)
sam.fineSplitFactor = self.fineSplitFactor
if self.triggers == 'mt': sam.triggers = mc_triggers_mt
if self.triggers == 'et': sam.triggers = mc_triggers_et
if self.triggers == 'tt': sam.triggers = mc_triggers_tt
if self.triggers == 'em': sam.triggers = mc_triggers_em
sam.triggers = data_triggers_2012C
elif '2012D' in sam.name:
sam.triggers = data_triggers_2012D
allsamples1 = copy.copy( mc_higgs )
allsamples1.extend( ztt_mc_ewk )
allsamples1.extend( embed_list )
allsamples1.extend( mc_higgs_susy )
pat1 = '%HTT_Feb6_hinzmann_Nom'
connect( allsamples1, pat1, 'diTau.*root', aliases, cache=True, verbose=False)
allsamples2 = copy.copy( data_2012 )
allsamples2.extend( t_mc_ewk )
allsamples2.extend( mc_diboson_incl )
allsamples2.extend( w_mc_ewk )
pat2 = '%V5_8_0/HTT_Sep12_hinzmann'
connect( allsamples2, pat2, 'diTau.*root', aliases, cache=True, verbose=False)
allsamples = copy.copy(allsamples1)
allsamples.extend(allsamples2)
DYJets.nGenEvents /= 0.8 ## overall trigger efficiency in MC
for h in mc_higgs :
h.nGenEvents /= 0.8 ## overall trigger efficiency in MC
for c in allsamples:
c.splitFactor = splitFactor(c)
# pat3 = '%HTT_22Jul_manzoni_Down'
pat3 = '%HTT_24Jul_newTES_manzoni_Down'
connect( allsamples0, pat0, 'diTau.*root', aliases, cache=True, verbose=False)
connect( allsamples1, pat1, 'diTau.*root', aliases, cache=True, verbose=False)
connect( allsamples2, pat2, 'diTau.*root', aliases, cache=True, verbose=False)
connect( allsamples3, pat3, 'diTau.*root', aliases, cache=True, verbose=False)
cache = True
T_tW.files = getFiles('/T_tW-channel-DR_TuneZ2star_8TeV-powheg-tauola/Summer12_DR53X-PU_S10_START53_V7A-v1/AODSIM/PAT_CMG_V5_16_0/HTT_22Jul_manzoni_Nom', 'manzoni', 'diTau_fullsel.*root', cache)
T_tW.nGenEvents = 497658.
#TTJetsHadronic.nGenEvents = 31223821.
#TTJetsHadronic.xSection = 249.5*0.676*0.676*0.96
TTJets_emb.files = getFiles('/TTJets_FullLeptMGDecays_8TeV-madgraph/StoreResults-Summer12_DR53X_PU_S10_START53_V7A_v1_ReplaceRecMuons_RHembedded_trans1_tau132_pthad1_30had2_30_v1-f456bdbb960236e5c696adfe9b04eaae/USER/V5_B/PAT_CMG_V5_16_0/HTT_22Jul_manzoni_Nom', 'manzoni', 'diTau_fullsel.*root', cache)
TTJets_emb.nGenEvents = 4243699.
TTJets_emb.xSection = 249.5 * (1-0.676)*(1-0.676)*0.96
allsamples = copy.copy( allsamples1 )
allsamples.extend( allsamples2 )
allsamples.extend( allsamples3 )
for c in allsamples:
c.splitFactor = int(100*splitFactor(c)/1)
# when stitching
WJets.nGenEvents += WJets_v2.nGenEvents
MC_list = copy.copy( mc_ewk )
MC_list.extend( mc_diboson_inc )
allsamples = copy.copy(MC_list)
allsamples.extend( data_list )
# the pattern provided to this function is a SQL pattern.
# % represents the wildcard.
# this pattern allows to select a whole group of samples.
connect( allsamples, '%CMGPF_V5_16_0', 'cmgTuple.*root', aliases, cache=True, verbose=False)
# guessing the best split factor for each component,
# to get about the same number of events in each job
for c in allsamples:
c.splitFactor = splitFactor(c, nEventsPerJob=2e5)
# picking up the json file automatically. Here, we use the dcs jsons.
for data in data_list:
if len(data.files):
data.json = jsonPick( data.files[0], jsonMap )
if __name__ == '__main__':
from CMGTools.RootTools.fwlite.Config import printComps
printComps(allsamples, True)
###################################################
### OPTIONS ###
###################################################
# Get all heppy options; set via "-o production" or "-o production=True"
# production = True run on batch, production = False (or unset) run locally
production = getHeppyOption('production' , False)
pick_events = getHeppyOption('pick_events', False)
###################################################
### HANDLE SAMPLES ###
###################################################
samples = [BdKstEE]
for sample in samples:
sample.triggers = ['HLT_DoubleMu3_Trk_Tau3mu_v%d' %i for i in range(3, 12)]
sample.splitFactor = splitFactor(sample, 2e4)
sample.puFileData = puFileData
sample.puFileMC = puFileMC
selectedComponents = samples
###################################################
### ANALYSERS ###
###################################################
eventSelector = cfg.Analyzer(
EventSelector,
name='EventSelector',
toSelect=[4148011548]
)
lheWeightAna = cfg.Analyzer(
'/WZJetsTo3LNu.*START42.*': 'WZJetsTo3LNu',
'/ZZJetsTo2L2Nu.*START42.*': 'ZZJetsTo2L2Nu',
'/ZZJetsTo2L2Q.*START42.*': 'ZZJetsTo2L2Q',
'/ZZJetsTo4L.*START42.*': 'ZZJetsTo4L',
}
MC_list = copy.copy(mc_ewk)
MC_list.extend(mc_higgs)
MC_list.extend(mc_diboson)
for sam in MC_list:
sam.triggers = mc_triggers
for data in data_list_2011A:
data.triggers = data_triggers_2011A
for data in data_list_2011B:
data.triggers = data_triggers_2011B
allsamples = copy.copy(MC_list)
allsamples.extend(data_list_2011)
allsamples.extend(embed_list_2011)
connect(allsamples,
'%TAUMU_2011_Colin_Oct24',
'tauMu.*root',
aliases,
cache=True,
verbose=False)
Tbar_tW.nGenEvents = 809984.
for c in allsamples:
c.splitFactor = splitFactor(c)
]
for dy in mc_dy:
dy.nevents = dy_nevents
# Attach number of generated events for stitching
w_nevents = [
WJets.nGenEvents, W1Jets.nGenEvents, W2Jets.nGenEvents, W3Jets.nGenEvents,
W4Jets.nGenEvents
]
for w in mc_w:
w.nevents = w_nevents
print[(s.name, s.dataset_entries) for s in allsamples if s.dataset_entries]
print[(s.name, s.dataset_entries) for s in allsamples if not s.dataset_entries]
for mc in MC_list:
mc.triggers = mc_triggers
# allsamples.append(mc)
for data in data_list:
if len(data.files):
data.json = jsonPick(data.files[0], jsonMap)
data.triggers = data_triggers
# allsamples.append(data)
for embed in embed_list:
if len(embed.files):
embed.json = jsonPick(embed.files[0], jsonMap)
embed.triggers = embed_triggers
# No trigger requirements for embedded samples
for c in allsamples:
c.splitFactor = splitFactor(c, 5e4)
# Add trigger corrections for the prompt lepton l0
MU_PROMPT_SFS = dc(MU_SFS)
MU_PROMPT_SFS['trigger'] = (SF_FILE, 'm_trg24_27_kit')
###################################################
### HANDLE SAMPLES ###
###################################################
samples = all_samples + all_signals
# samples = [TTJets, TTJets_ext]
###################################################
# FIXME! are trigger names and filters correct regardless of the year?
# triggers same for 2018: https://tomc.web.cern.ch/tomc/triggerPrescales/2018//?match=Ele
for sample in samples:
sample.triggers = triggers_ele_mc + triggers_mu_mc
sample.splitFactor = splitFactor(sample, 7e5)
if sample in all_signals:
sample.splitFactor = splitFactor(sample, 5e5)
sample.puFileMC = '$CMSSW_BASE/src/CMGTools/HNL/data/pileup/mc/2018/MC_PileUp_2018_Autumn18.root'
sample.puFileData = '$CMSSW_BASE/src/CMGTools/HNL/data/pileup/data/2018/Data_PileUp_2018_69p2.root'
selectedComponents = samples
###################################################
# set to True if you want to run interactively on a selected portion of samples/files/whatnot
testing = True
if testing:
# run on a single component
# comp = samples[0]
# comp.files = comp.files[:1]
n_events_per_job = 1e6
if test:
n_events_per_job = 1e5
if embedded:
n_events_per_job = 3e4
for sample in selectedComponents:
if data:
sample.triggers = data_triggers
sample.triggerobjects = data_triggerfilters
if embedded:
sample.triggerobjects = embedded_triggerfilters
else:
sample.triggers = mc_triggers
sample.triggerobjects = mc_triggerfilters
sample.splitFactor = splitFactor(sample, n_events_per_job)
sample.channel = 'tt'
if test:
cache = True
selectedComponents = [selectedComponents[0]]
for comp in selectedComponents:
comp.files = comp.files[:1]
comp.splitFactor = 1
comp.fineSplitFactor = 1
events_to_pick = []
#KIT's skimming function
def skim_KIT(event):
'/DoubleMu/Run2011B-16Jan2012-v1/AOD/V5/PAT_CMG_V5_6_0_B$':'data_Run2011B',
'/DoubleMu/Run2011A-16Jan2012-v1/AOD/V5/PAT_CMG_V5_6_0_B$':'data_Run2011A',
'/TTJets.*START42.*/VertexWeight':'TTJets',
'/WWJetsTo2L2Nu.*START42.*/VertexWeight':'WWJetsTo2L2Nu',
'/WZJetsTo2L2Q.*START42.*/VertexWeight':'WZJetsTo2L2Q',
'/WZJetsTo3LNu.*START42.*/VertexWeight':'WZJetsTo3LNu',
'/ZZJetsTo2L2Nu.*START42.*/VertexWeight':'ZZJetsTo2L2Nu',
'/ZZJetsTo2L2Q.*START42.*/VertexWeight':'ZZJetsTo2L2Q',
'/ZZJetsTo4L.*START42.*/VertexWeight':'ZZJetsTo4L',
}
data_list = data_list_2011
MC_list = copy.copy( mc_ewk )
MC_list.extend( mc_diboson )
allsamples = copy.copy(MC_list)
allsamples.extend( data_list_2011 )
connect( allsamples, '%PAT_CMG_V5_6_0_B%', 'cmgTuple.*root', aliases, cache=True, verbose=False)
for c in allsamples:
c.splitFactor = splitFactor(c, nEventsPerJob=2e5)
for data in data_list:
if len(data.files):
data.json = jsonPick( data.files[0])
WJetsToLNu, WJetsToLNu_ext, TTJets
]
#samples = all_signals_m
auxsamples = [] #[ttbar, DYJetsToLL_M50]
# samples = [comp for comp in samples if comp.name=='TTJets_amcat']
for sample in samples + auxsamples:
sample.triggers = ['HLT_IsoMu24_v%d' % i
for i in range(1, 15)] #muon trigger
sample.triggers += ['HLT_IsoMu27_v%d' % i
for i in range(1, 15)] #muon trigger
sample.triggers += ['HLT_Mu50_v%d' % i
for i in range(1, 15)] #muon trigger
sample.splitFactor = splitFactor(sample, 1e5)
selectedComponents = samples
###################################################
### ANALYSERS ###
###################################################
eventSelector = cfg.Analyzer(EventSelector,
name='EventSelector',
toSelect=[326])
lheWeightAna = cfg.Analyzer(LHEWeightAnalyzer,
name="LHEWeightAnalyzer",
useLumiInfo=False)
jsonAna = cfg.Analyzer(
###################################################
### OPTIONS ###
###################################################
# Get all heppy options; set via "-o production" or "-o production=True"
# production = True run on batch, production = False (or unset) run locally
production = getHeppyOption('production', False)
pick_events = getHeppyOption('pick_events', False)
###################################################
### HANDLE SAMPLES ###
###################################################
samples = [HN3L_M_7_V_0p00547722557505_mu_onshell]
for sample in samples:
sample.triggers = ['HLT_IsoMu24_v%d' % i for i in range(4, 5)]
sample.splitFactor = splitFactor(sample, nEventsPerJob=1e5)
sample.puFileData = puFileData
sample.puFileMC = puFileMC
selectedComponents = samples
###################################################
### ANALYSERS ###
###################################################
eventSelector = cfg.Analyzer(EventSelector, name='EventSelector', toSelect=[])
lheWeightAna = cfg.Analyzer(LHEWeightAnalyzer,
name="LHEWeightAnalyzer",
useLumiInfo=False)
jsonAna = cfg.Analyzer(
def generateKeyConfigs(samples,production, promptLeptonType, L1L2LeptonType, isData, isSignal):
import os
from collections import OrderedDict
import PhysicsTools.HeppyCore.framework.config as cfg
from PhysicsTools.HeppyCore.framework.config import printComps
from PhysicsTools.HeppyCore.framework.heppy_loop import getHeppyOption
from PhysicsTools.Heppy.utils.cmsswPreprocessor import CmsswPreprocessor
from CMGTools.RootTools.utils.splitFactor import splitFactor
# import Heppy analyzers:
from PhysicsTools.Heppy.analyzers.core.JSONAnalyzer import JSONAnalyzer
from PhysicsTools.Heppy.analyzers.core.SkimAnalyzerCount import SkimAnalyzerCount
from PhysicsTools.Heppy.analyzers.core.EventSelector import EventSelector
from PhysicsTools.Heppy.analyzers.objects.VertexAnalyzer import VertexAnalyzer
from PhysicsTools.Heppy.analyzers.core.PileUpAnalyzer import PileUpAnalyzer
from PhysicsTools.Heppy.analyzers.gen.GeneratorAnalyzer import GeneratorAnalyzer
from PhysicsTools.Heppy.analyzers.gen.LHEWeightAnalyzer import LHEWeightAnalyzer
from CMGTools.H2TauTau.proto.analyzers.TriggerAnalyzer import TriggerAnalyzer
# import HNL analyzers:
from CMGTools.HNL.analyzers.HNLAnalyzer import HNLAnalyzer
from CMGTools.HNL.analyzers.HNLTreeProducer import HNLTreeProducer
from CMGTools.HNL.analyzers.HNLGenTreeAnalyzer import HNLGenTreeAnalyzer
from CMGTools.HNL.analyzers.RecoGenAnalyzer import RecoGenAnalyzer
from CMGTools.HNL.analyzers.TriggerAnalyzer import TriggerAnalyzer
from CMGTools.HNL.analyzers.JetAnalyzer import JetAnalyzer
from CMGTools.HNL.analyzers.METFilter import METFilter
from CMGTools.HNL.analyzers.LeptonWeighter import LeptonWeighter
from pdb import set_trace
###################################################
### OPTIONS ###
###################################################
# Get all heppy options; set via "-o production" or "-o production=True"
# production = True run on batch, production = False (or unset) run locally
production = getHeppyOption('production' , production)
pick_events = getHeppyOption('pick_events', False)
promptLeptonType = promptLeptonType # choose from 'ele' or 'mu'
L1L2LeptonType = L1L2LeptonType #choose from 'ee', 'mm', 'em'
###################################################
### HANDLE SAMPLES ###
###################################################
#samples = [comp for comp in samples if comp.name=='TTJets_amcat']
if promptLeptonType == 'ele':
for sample in samples:
sample.triggers = ['HLT_Ele27_WPTight_Gsf_v%d' %i for i in range(1, 15)] #electron trigger
sample.triggers += ['HLT_Ele32_WPTight_Gsf_v%d' %i for i in range(1, 15)] #electron trigger
sample.triggers += ['HLT_Ele35_WPTight_Gsf_v%d' %i for i in range(1, 15)] #electron trigger
sample.triggers += ['HLT_Ele115_CaloIdVT_GsfTrkIdT_v%d' %i for i in range(1, 15)] #electron trigger
sample.triggers += ['HLT_Ele135_CaloIdVT_GsfTrkIdT_v%d' %i for i in range(1, 15)] #electron trigger
sample.splitFactor = splitFactor(sample, 1e5)
if promptLeptonType == 'mu':
for sample in samples:
sample.triggers = ['HLT_IsoMu24_v%d' %i for i in range(1, 15)] #muon trigger
sample.triggers += ['HLT_IsoMu27_v%d' %i for i in range(1, 15)] #muon trigger
sample.triggers += ['HLT_Mu50_v%d' %i for i in range(1, 15)] #muon trigger
sample.splitFactor = splitFactor(sample, 1e5)
selectedComponents = samples
###################################################
### ANALYZERS ###
###################################################
eventSelector = cfg.Analyzer(
EventSelector,
name='EventSelector',
toSelect=[326]
)
lheWeightAna = cfg.Analyzer(
LHEWeightAnalyzer, name="LHEWeightAnalyzer",
useLumiInfo=False
)
jsonAna = cfg.Analyzer(
JSONAnalyzer,
name='JSONAnalyzer',
)
skimAna = cfg.Analyzer(
SkimAnalyzerCount,
name='SkimAnalyzerCount'
)
triggerAna = cfg.Analyzer(
TriggerAnalyzer,
name='TriggerAnalyzer',
addTriggerObjects=True,
requireTrigger=True,
usePrescaled=False,
unpackLabels=True,
)
vertexAna = cfg.Analyzer(
VertexAnalyzer,
name='VertexAnalyzer',
fixedWeight=1,
keepFailingEvents=False,
verbose=False
)
pileUpAna = cfg.Analyzer(
PileUpAnalyzer,
name='PileUpAnalyzer',
true=True
)
metFilter = cfg.Analyzer(
METFilter,
name='METFilter',
processName='PAT', #mschoene: Filters very much do exist in MC and most of them should be applied to MC as well, but not all!
fallbackProcessName = 'RECO',
triggers=[
'Flag_goodVertices',
'Flag_globalSuperTightHalo2016Filter',
'Flag_HBHENoiseFilter',
'Flag_HBHENoiseIsoFilter',
'Flag_EcalDeadCellTriggerPrimitiveFilter',
'Flag_BadPFMuonFilter',
'Flag_BadChargedCandidateFilter',
'Flag_eeBadScFilter',
'Flag_ecalBadCalibFilter',
]
)
HNLTreeProducer = cfg.Analyzer(
HNLTreeProducer,
name='HNLTreeProducer',
L1L2LeptonType=L1L2LeptonType,
promptLepType=promptLeptonType,
)
HNLGenTreeAnalyzer = cfg.Analyzer(
HNLGenTreeAnalyzer,
name='HNLGenTreeAnalyzer',
)
RecoGenAnalyzer = cfg.Analyzer(
RecoGenAnalyzer,
name='RecoGenAnalyzer',
)
genAna = GeneratorAnalyzer.defaultConfig
genAna.allGenTaus = True # save in event.gentaus *ALL* taus, regardless whether hadronic / leptonic decay
# for each path specify which filters you want the electrons/muons to match to
triggers_and_filters = OrderedDict()
if promptLeptonType == 'ele':
triggers_and_filters['HLT_Ele27_WPTight_Gsf'] = 'hltEle27WPTightGsfTrackIsoFilter'
triggers_and_filters['HLT_Ele32_WPTight_Gsf'] = 'hltEle32WPTightGsfTrackIsoFilter'
triggers_and_filters['HLT_Ele35_WPTight_Gsf'] = 'hltEle35noerWPTightGsfTrackIsoFilter'
triggers_and_filters['HLT_Ele115_CaloIdVT_GsfTrkIdT'] = 'hltEle115CaloIdVTGsfTrkIdTGsfDphiFilter'
triggers_and_filters['HLT_Ele135_CaloIdVT_GsfTrkIdT'] = 'hltEle135CaloIdVTGsfTrkIdTGsfDphiFilter'
if promptLeptonType == 'mu':
triggers_and_filters['HLT_IsoMu24'] = 'hltL3crIsoL1sSingleMu22L1f0L2f10QL3f24QL3trkIsoFiltered0p07'
triggers_and_filters['HLT_IsoMu27'] = 'hltL3crIsoL1sMu22Or25L1f0L2f10QL3f27QL3trkIsoFiltered0p07'
triggers_and_filters['HLT_Mu50'] = 'hltL3fL1sMu22Or25L1f0L2f10QL3Filtered50Q'
HNLAnalyzer = cfg.Analyzer(
HNLAnalyzer,
name='HNLAnalyzer',
promptLepton=promptLeptonType,
L1L2LeptonType=L1L2LeptonType,
triggersAndFilters=triggers_and_filters,
candidate_selection='maxpt',
)
if promptLeptonType == 'ele':
Weighter_l0 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l0',
scaleFactorFiles={
'trigger' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_trg_SingleEle_Ele32OREle35_desy'),
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l0(),
disable=False
)
if promptLeptonType == 'mu':
Weighter_l0 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l0',
scaleFactorFiles={
'trigger' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_trg_SingleMu_Mu24ORMu27_desy'),
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l0(),
disable=False
)
if L1L2LeptonType == 'mm':
Weighter_l1 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l1',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l1(),
disable=True
)
Weighter_l2 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l2',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l2(),
disable=True
)
if L1L2LeptonType == 'em':
Weighter_l1 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l1',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l1(),
disable=True
)
Weighter_l2 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l2',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'm_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l2(),
disable=True
)
if L1L2LeptonType == 'ee':
Weighter_l1 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l1',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l1(),
disable=True
)
Weighter_l2 = cfg.Analyzer(
LeptonWeighter,
name='LeptonWeighter_l2',
scaleFactorFiles={
'idiso' :('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_id'),
'tracking':('$CMSSW_BASE/src/CMGTools/HNL/data/leptonsf/htt_scalefactors_v17_1.root', 'e_iso'),
},
dataEffFiles={
# 'trigger':('$CMSSW_BASE/src/CMGTools/H2TauTau/data/htt_scalefactors_v16_2.root', 'm_trgIsoMu22orTkIsoMu22_desy'),
},
getter = lambda event : event.the_3lep_cand.l2(),
disable=True
)
# see SM HTT TWiki
# https://twiki.cern.ch/twiki/bin/viewauth/CMS/SMTauTau2016#Jet_Energy_Corrections
jetAna = cfg.Analyzer(
JetAnalyzer,
name = 'JetAnalyzer',
jetCol = 'slimmedJets',
jetPt = 20.,
jetEta = 5.,
relaxJetId = False, # relax = do not apply jet ID
relaxPuJetId = True, # relax = do not apply pileup jet ID
jerCorr = False,
puJetIDDisc = 'pileupJetId:fullDiscriminant',
recalibrateJets = False,
applyL2L3Residual = 'MC',
# RM: FIXME! check the GTs
# mcGT = '94X_mc2017_realistic_v14',
# dataGT = '94X_dataRun2_v6',
#jesCorr = 1., # Shift jet energy scale in terms of uncertainties (1 = +1 sigma)
)
###################################################
### SEQUENCE ###
###################################################
if isData == True:
sequence = cfg.Sequence([
# eventSelector,
jsonAna,
skimAna,
triggerAna,
vertexAna,
pileUpAna,
HNLAnalyzer,
jetAna,
metFilter,
HNLTreeProducer,
])
if isData == False:
if isSignal == True:
sequence = cfg.Sequence([
# eventSelector,
lheWeightAna, # les houches
jsonAna,
skimAna,
triggerAna,
vertexAna,
pileUpAna,
genAna,
HNLGenTreeAnalyzer,
# RecoGenAnalyzer,
HNLAnalyzer,
Weighter_l0,
Weighter_l1,
Weighter_l2,
jetAna,
metFilter,
HNLTreeProducer,
])
if isSignal == False:
sequence = cfg.Sequence([
# eventSelector,
lheWeightAna, # les houches
jsonAna,
skimAna,
triggerAna,
vertexAna,
pileUpAna,
genAna,
HNLGenTreeAnalyzer,
HNLAnalyzer,
Weighter_l0,
Weighter_l1,
Weighter_l2,
jetAna,
metFilter,
HNLTreeProducer,
])
###################################################
### SET BATCH OR LOCAL ###
###################################################
if not production:
# comp = HN3L_M_2p5_V_0p0173205080757_e_onshell
# comp = HN3L_M_2p5_V_0p00707106781187_e_onshell
# comp = all_signals_e[0]
# comp = DYJetsToLL_M50
comp = samples[0]
# comp = samples
# comp = ttbar
# comp = bkg
selectedComponents = [comp]
comp.splitFactor = 1
comp.fineSplitFactor = 1
comp.files = comp.files[:1] #mschoene: for testing it's much quicker to run just on the first file
###################################################
### PREPROCESSOR ###
###################################################
preprocessor = True #mschoene: for false it will use EOSEventsWithDownload, which actually copies the file, which is extremely slow
#temporary copy remote files using xrd
from PhysicsTools.HeppyCore.framework.eventsfwlite import Events
from CMGTools.HNL.utils.EOSEventsWithDownload import EOSEventsWithDownload
event_class = EOSEventsWithDownload if not preprocessor else Events
EOSEventsWithDownload.aggressive = 2 # always fetch if running on Wigner
EOSEventsWithDownload.long_cache = getHeppyOption('long_cache', False)
# if preprocessor: preprocessor.prefetch = prefetch #mschoene: removed as unused
# if extrap_muons_to_L1:
# fname = '$CMSSW_BASE/src/CMGTools/WTau3Mu/prod/muon_extrapolator_cfg.py'
# sequence.append(fileCleaner)
# preprocessor = CmsswPreprocessor(fname, addOrigAsSecondary=False)
# if compute_mvamet:
# fname = '$CMSSW_BASE/src/CMGTools/WTau3Mu/prod/compute_mva_met_data_cfg.py'
# sequence.append(fileCleaner)
# preprocessor = CmsswPreprocessor(fname, addOrigAsSecondary=False)
# the following is declared in case this cfg is used in input to the heppy.py script
config = cfg.Config(
components = selectedComponents,
sequence = sequence,
services = [],
# preprocessor = preprocessor, #mschoene: removed as unused
events_class = event_class
)
printComps(config.components, True)
return config
sequence.append(treeProducer)
if syncntuple:
sequence.append(syncTreeProducer)
###################################################
# Minimal list of samples
if computeSVfit:
split_factor = 5e3
for sample in samples:
sample.triggers = mc_triggers
sample.triggerobjects = mc_triggerfilters
sample.splitFactor = splitFactor(sample, nevts_per_file)
sample.puFileData = puFileData
sample.puFileMC = puFileMC
for sample in data_list:
sample.triggers = data_triggers
sample.triggerobjects = data_triggerfilters
sample.splitFactor = splitFactor(sample, split_factor)
# Samples to be processed
selectedComponents = samples
if pick_events:
eventSelector.toSelect = [71838, 55848]
sequence.insert(0, eventSelector)
if cmssw:
tauMuAna.from_single_objects = False
# Minimal list of samples
samples = backgrounds_mu + sm_signals + sync_list + mssm_signals
split_factor = 1e5
if computeSVfit:
split_factor = 5e3
for sample in samples:
sample.triggers = mc_triggers
sample.triggerobjects = mc_triggerfilters
sample.splitFactor = splitFactor(sample, split_factor)
sample.puFileData = puFileData
sample.puFileMC = puFileMC
data_list = data_single_muon
for sample in data_list:
sample.triggers = data_triggers
sample.triggerobjects = data_triggerfilters
sample.splitFactor = splitFactor(sample, split_factor)
sample.json = '/afs/cern.ch/cms/CAF/CMSCOMM/COMM_DQM/certification/Collisions15/13TeV/Cert_246908-260627_13TeV_PromptReco_Collisions15_25ns_JSON_v2.txt'
sample.lumi = 2260.
###################################################
### SET COMPONENTS BY HAND ###
"/Tau/Run2015B-PromptReco-v1/MINIAOD",
"CMS",
".*root",
"/afs/cern.ch/cms/CAF/CMSCOMM/COMM_DQM/certification/Collisions15/13TeV/DCSOnly/json_DCSONLY_Run2015B.txt"
)
MC_list = [ggh160]
run2015B.intLumi = '2.0' # in pb
run2015B.triggers = mc_triggers_tt
split_factor = 1e5
for sample in MC_list:
sample.triggers = mc_triggers_tt
sample.splitFactor = splitFactor(sample, split_factor)
data_list = [run2015B]
###################################################
### ASSIGN PU to MC ###
###################################################
for mc in MC_list:
mc.puFileData = puFileData
mc.puFileMC = puFileMC
###################################################
### SET COMPONENTS BY HAND ###
###################################################
selectedComponents = MC_list #+ data_list
# selectedComponents = mc_dict['HiggsGGH125']
# from CMGTools.Production.getFiles import getFiles
from CMGTools.RootTools.utils.connect import connect
from CMGTools.RootTools.utils.splitFactor import splitFactor
from CMGTools.H2TauTau.proto.samples.phys14.higgs import mc_higgs
from CMGTools.H2TauTau.proto.samples.phys14.triggers_tauMu import mc_triggers
aliases = {
'/GluGluToHToTauTau.*Phys14DR.*' : 'HiggsGGH',
'/VBF_HToTauTau_M-125_13TeV-powheg-pythia6.*Phys14DR.*' : 'HiggsVBF',
}
# dictionarize
mc_dict = {}
for s in mc_higgs:
mc_dict[s.name] = s
MC_list = [v for k, v in mc_dict.items()]#[mc_dict['HiggsGGH125'], mc_dict['HiggsVBF125']]
for sam in MC_list:
# Set MC triggers
sam.triggers = mc_triggers
connect(MC_list, '%TAUMU_MINIAOD_SVFIT25ns_steggema', '.*root', aliases, cache=True, verbose=False)
MC_list = [m for m in MC_list if m.files]
for sample in MC_list:
sample.splitFactor = splitFactor(sample, 10e4)
