def main():
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
sys.exit(0)
else:
pass
###################
## SETUP THE PROCESS
###################
maxEvents = {}
process = Process(prefix, maxEvents=maxEvents)
###################
## ADD DATASETS
###################
process.addDatasetsFromMulticrab(sys.argv[1],
blacklist=blacklist,
whitelist=whitelist)
# Enable genuine tau histograms for common plots (needed for calculating N_QCD)
allSelections.CommonPlots.enableGenuineTauHistograms = True
# Set splitting of phase space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
PSet(label="tauPt_1", binLowEdges=[40, 60], useAbsoluteValues=False),
PSet(label="decayMode_1", binLowEdges=[2, 3], useAbsoluteValues=False),
PSet(label="tauPt_2",
binLowEdges=[20, 40, 60],
useAbsoluteValues=False),
PSet(label="decayMode_2",
binLowEdges=[1, 2, 3],
useAbsoluteValues=False),
]
##################
## BUILD ANALYSIS MODULES
###################
builder = AnalysisBuilder(
prefix,
dataEras,
searchModes,
### OPRIONS ###
usePUreweighting=True,
useTopPtReweighting=False,
doSystematicVariations=True,
analysisType="HToHW_background")
builder.build(process, allSelections)
process.run()
def _setupQGL(jsonname):
# Read json
_jsonpath = os.path.join(os.getenv("HIGGSANALYSIS_BASE"), "NtupleAnalysis",
"data", "QGLR")
filename = os.path.join(_jsonpath, jsonname)
if not os.path.exists(filename):
raise Exception("Error: file '%s' does not exist!" % filename)
f = open(filename)
contents = json.load(f)
f.close()
# Loop over the contents to convert as list of PSets the requested information
psetList = []
for row in contents:
p = PSet(
jetType=row["Jet"],
prob=float(row["prob"]),
probError=float(row["probError"]),
QGLmin=float(row["QGLmin"]),
QGLmax=float(row["QGLmax"]),
Ptmin=float(row["Ptmin"]),
Ptmax=float(row["Ptmax"]),
)
psetList.append(p)
return psetList
def _setupBtagEfficiency(btagPset, btagEfficiencyFilename, direction,
variationInfo):
fullname = os.path.join(os.getenv("HIGGSANALYSIS_BASE"), "NtupleAnalysis",
"data", btagEfficiencyFilename)
if not os.path.exists(fullname):
raise Exception(
"Error: Could not find the btag efficiency json file! (tried: %s)"
% fullname)
# Read the json file
f = open(fullname)
contents = json.load(f)
f.close()
# Loop over the contents to convert as list of PSets the requested information
psetList = []
for row in contents:
# Require that the btag discriminator and working points match
if row["discr"] == btagPset.__getattr__(
"bjetDiscr") and row["workingPoint"] == btagPset.__getattr__(
"bjetDiscrWorkingPoint"):
#print row["discr"], row["workingPoint"], row["flavor"], row["ptMin"]
p = PSet(jetFlavor=row["flavor"],
ptMin=row["ptMin"],
ptMax=row["ptMax"],
eff=float(row["eff"]),
effDown=float(row["effDown"]),
effUp=float(row["effUp"]))
psetList.append(p)
btagPset.btagEfficiency = psetList
def createAnalyzer(dataVersion, era):
a = Analyzer(
"TriggerEfficiency",
name=era,
Trigger=PSet(triggerOR=[], triggerOR2=[]),
usePileupWeights=True,
offlineSelection=leg,
MuonSelection=PSet(
# discriminators = ["muIDMedium"],
# discriminators = ["TrgMatch_IsoMu20_eta2p1"],
# discriminators = ["Muons_TrgMatch_IsoMu16_eta2p1"],
),
TauSelection=PSet(
discriminators=[
"byLooseCombinedIsolationDeltaBetaCorr3Hits", #"byMediumIsolationMVA3newDMwLT",
"againstMuonTight3",
"againstElectronMediumMVA6"
],
nprongs=1,
relaxedOfflineSelection=False,
),
binning=binning,
xLabel=xLabel,
yLabel=yLabel,
)
if isData(dataVersion):
if "2016" in era:
a.Trigger.triggerOR = ["HLT_IsoMu22_eta2p1_vx"]
a.Trigger.triggerOR2 = [
"HLT_VLooseIsoPFTau" + tauThreshold + "_Trk50_eta2p1_vx"
]
if tauThreshold == "50":
a.Trigger.triggerOR2 = ["HLT_LooseIsoPFTau50_Trk30_eta2p1_vx"]
a.runMin = runmin
a.runMax = runmax
else:
if "2016" in era:
a.Trigger.triggerOR = ["HLT_IsoMu22_eta2p1_vx"]
a.Trigger.triggerOR2 = [
"HLT_VLooseIsoPFTau" + tauThreshold + "_Trk50_eta2p1_vx"
]
if tauThreshold == "50":
a.Trigger.triggerOR2 = ["HLT_LooseIsoPFTau50_Trk30_eta2p1_vx"]
return a
def pileupWeight(data=None, mc=None, enabled=None):
if data is not None and mc is not None and enabled is None:
enabled = True
if enabled is None:
enabled = False
pset = PSet(enabled=enabled)
if not enabled:
return pset
if data is None:
raise Exception("If pileupWeight is enabled, must give parameter 'data' for the data era")
if mc is None:
raise Exception("If pileupWeight is enabled, must give parameter 'mc' for the MC era")
pset.data = File(os.path.join(_pileupHistogramPath, "PileupHistogramData"+data+".root"))
pset.mc = File(os.path.join(_pileupHistogramPath, "PileupHistogramMC"+mc+".root"))
return pset
def _assignTrgSF(name, binEdges, SF, SFup, SFdown, pset, direction):
if not direction in ["nominal", "up", "down"]:
raise Exception("Error: unknown option for SF direction('%s')!" %
direction)
myScaleFactors = SF[:]
if direction == "up":
myScaleFactors = SFup[:]
elif direction == "down":
myScaleFactors = SFdown[:]
setattr(pset, name,
PSet(binLeftEdges=binEdges[:], scaleFactors=myScaleFactors))
def pileupWeight(data=None, mc=None, enabled=None):
if data is not None and mc is not None and enabled is None:
enabled = True
if enabled is None:
enabled = False
pset = PSet(enabled=enabled)
if not enabled:
return pset
if data is None:
raise Exception(
"If pileupWeight is enabled, must give parameter 'data' for the data era"
)
if mc is None:
raise Exception(
"If pileupWeight is enabled, must give parameter 'mc' for the MC era"
)
pset.data = File(
os.path.join(_pileupHistogramPath,
"PileupHistogramData" + data + ".root"))
pset.mc = File(
os.path.join(_pileupHistogramPath, "PileupHistogramMC" + mc + ".root"))
return pset
def main():
# Save start time (epoch seconds)
tStart = time.time()
Verbose("Started @ " + str(tStart), True)
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
print __doc__
sys.exit(0)
else:
pass
# ================================================================================================
# Setup the process
# ================================================================================================
maxEvents = {}
maxEvents["All"] = opts.nEvts
# maxEvents["2016"] = 1
# maxEvents["ZZTo4Q"] = -1
# maxEvents["ZJetsToQQ_HT600toInf"] = 1
# maxEvents["WZ_ext1"] = 1
# maxEvents["WZ"] = 1
# maxEvents["WWTo4Q"] = 1
# maxEvents["WJetsToQQ_HT_600ToInf"] = 1
# maxEvents["TTZToQQ"] = 1
# maxEvents["TTWJetsToQQ"] = 1
# maxEvents["TTTT"] = 1
# maxEvents["TT"] = 1
# maxEvents["ST_t_channel_top_4f_inclusiveDecays"] = 1
# maxEvents["ST_t_channel_antitop_4f_inclusiveDecays"] = 1
# maxEvents["ST_tW_top_5f_inclusiveDecays_ext1"] = 1
# maxEvents["ST_tW_top_5f_inclusiveDecays"] = 1
# maxEvents["ST_tW_antitop_5f_inclusiveDecays_ext1"] = 1
# maxEvents["ST_tW_antitop_5f_inclusiveDecays"] = 1
# maxEvents["ST_s_channel_4f_InclusiveDecays"] = 1
# maxEvents["QCD_HT700to1000_ext1"] = 1
# maxEvents["QCD_HT700to1000"] = 1
# maxEvents["QCD_HT50to100"] = 1
# maxEvents["QCD_HT500to700_ext1"] = 1
# maxEvents["QCD_HT500to700"] = 1
# maxEvents["QCD_HT300to500_ext1"] = 1
# maxEvents["QCD_HT300to500"] = 1
# maxEvents["QCD_HT200to300_ext1"] = 1
# maxEvents["QCD_HT200to300"] = 1
# maxEvents["QCD_HT2000toInf_ext1"] = 1
# maxEvents["QCD_HT2000toInf"] = 1
# maxEvents["QCD_HT1500to2000_ext1"] = 1
# maxEvents["QCD_HT1500to2000"] = 1
# maxEvents["QCD_HT100to200"] = 1
# maxEvents["QCD_HT1000to1500_ext1"] = 1
# maxEvents["QCD_HT1000to1500"] = 1
# maxEvents["JetHT_Run2016H_03Feb2017_ver3_v1_284036_284044"] = 1
# maxEvents["JetHT_Run2016H_03Feb2017_ver2_v1_281613_284035"] = 1
# maxEvents["JetHT_Run2016G_03Feb2017_v1_278820_280385"] = 1
# maxEvents["JetHT_Run2016F_03Feb2017_v1_278801_278808"] = 1
# maxEvents["JetHT_Run2016F_03Feb2017_v1_277932_278800"] = 1
# maxEvents["JetHT_Run2016E_03Feb2017_v1_276831_277420"] = 1
# maxEvents["JetHT_Run2016D_03Feb2017_v1_276315_276811"] = 1
# maxEvents["JetHT_Run2016C_03Feb2017_v1_275656_276283"] = 1
# maxEvents["JetHT_Run2016B_03Feb2017_ver2_v2_273150_275376"] = 1
# maxEvents["DYJetsToQQ_HT180"] = 1
# maxEvents["ChargedHiggs_HplusTB_HplusToTB_M_500"] = 1
process = Process(prefix, postfix, maxEvents)
# ================================================================================================
# Add the datasets (according to user options)
# ================================================================================================
if (opts.includeOnlyTasks):
Verbose("Adding only dataset %s from multiCRAB directory %s" %
(opts.includeOnlyTasks, opts.mcrab))
process.addDatasetsFromMulticrab(
opts.mcrab, includeOnlyTasks=opts.includeOnlyTasks)
elif (opts.excludeTasks):
Verbose("Adding all datasets except %s from multiCRAB directory %s" %
(opts.excludeTasks, opts.mcrab))
Print(
"If collision data are present, then vertex reweighting is done according to the chosen data era (era=2015C, 2015D, 2015) etc..."
)
process.addDatasetsFromMulticrab(opts.mcrab,
excludeTasks=opts.excludeTasks)
else:
myBlackList = [
"M_180", "M_200", "M_220", "M_250", "M_300", "M_350", "M_400",
"M_500", "M_650", "M_800", "M_1000", "M_1500", "M_2000", "M_2500",
"M_3000", "M_5000", "M_7000", "M_10000", "QCD"
]
Verbose("Adding all datasets from multiCRAB directory %s except %s" %
(opts.mcrab, (",".join(myBlackList))))
Verbose(
"Vertex reweighting is done according to the chosen data era (%s)"
% (",".join(dataEras)))
# process.addDatasetsFromMulticrab(opts.mcrab, blacklist=myBlackList)
if len(myBlackList) > 0:
regex = "|".join(myBlackList)
process.addDatasetsFromMulticrab(opts.mcrab, excludeTasks=regex)
else:
process.addDatasetsFromMulticrab(opts.mcrab)
# ================================================================================================
# Overwrite Default Settings
# ================================================================================================
from HiggsAnalysis.NtupleAnalysis.parameters.hplus2tbAnalysis import allSelections
allSelections.verbose = opts.verbose
allSelections.histogramAmbientLevel = opts.histoLevel
# Set splitting of phase-space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
# PSet(label="TetrajetBjetPt" , binLowEdges=[60, 100], useAbsoluteValues=False),
# PSet(label="TetrajetBjetEta", binLowEdges=[0.8, 1.6, 2.1], useAbsoluteValues=True), # B) not bad for -1.0 < BDT < 0.4
## PSet(label="TetrajetBjetPt" , binLowEdges=[120], useAbsoluteValues=False),
## PSet(label="TetrajetBjetEta", binLowEdges=[0.6, 1.2, 1.8, 2.1], useAbsoluteValues=True),
## PSet(label="TetrajetBjetPt" , binLowEdges=[100], useAbsoluteValues=False),
## PSet(label="TetrajetBjetEta", binLowEdges=[0.6, 0.9, 1.2, 1.5, 1.8, 2.1], useAbsoluteValues=True),
## PSet(label="TetrajetBjetPt" , binLowEdges=[80], useAbsoluteValues=False),
## PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 0.8, 1.6, 2.0], useAbsoluteValues=True),
### Default binning
#PSet(label="TetrajetBjetPt" , binLowEdges=[60, 80, 160], useAbsoluteValues=False), # Alexandros (40-60, and 60-80 bins off. others v. good)
#PSet(label="TetrajetBjetEta", binLowEdges=[1.0, 1.8], useAbsoluteValues=True), # Alexandros good!
#PSet(label="TetrajetBjetPt" , binLowEdges=[80, 160], useAbsoluteValues=False), # Alexandros (40-60, and 60-80 bins off. others good)
#PSet(label="TetrajetBjetEta", binLowEdges=[1.0, 1.8], useAbsoluteValues=True), # Alexandros good!
#
# PSet(label="TetrajetBjetPt" , binLowEdges=[80, 200], useAbsoluteValues=False), # Fotis
# PSet(label="TetrajetBjetEta", binLowEdges=[0.8, 1.6], useAbsoluteValues=True), # Fotis
#
# PSet(label="TetrajetBjetPt" , binLowEdges=[60, 80, 140, 200], useAbsoluteValues=False), #Fotis
# PSet(label="TetrajetBjetEta", binLowEdges=[0.8, 1.5, 2.1], useAbsoluteValues=True), #Fotis
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 60, 80, 100, 160], useAbsoluteValues=False),
# PSet(label="TetrajetBjetEta", binLowEdges=[0.2, 0.4, 0.6, 0.8, 1.6, 2.0, 2.2], useAbsoluteValues=True),
PSet(label="TetrajetBjetEta",
binLowEdges=[0.4, 0.8, 1.6, 2.0, 2.2],
useAbsoluteValues=True), #AN v4
### Other attempts
# PSet(label="TetrajetBjetEta", binLowEdges=[-2.2, -2.0, -1.6, -0.8, -0.4, +0.4, +0.8, +1.6, +2.0, +2.2], useAbsoluteValues=False),
# PSet(label="TetrajetBjetPt" , binLowEdges=[100], useAbsoluteValues=False), # C)
# PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 1.2, 1.8, 2.1], useAbsoluteValues=True), # C)
# PSet(label="TetrajetBjetPt" , binLowEdges=[60, 100], useAbsoluteValues=False), # B) not bad for -1.0 < BDT < 0.4
# PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 1.2, 1.8, 2.0], useAbsoluteValues=True), # B) not bad for -1.0 < BDT < 0.4
# PSet(label="TetrajetBjetPt" , binLowEdges=[120, 200], useAbsoluteValues=False), # A) not great for -1.0 < BDT < 0.4
# PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 1.2, 1.8, 2.1], useAbsoluteValues=True), # A) not great for -1.0 < BDT < 0.4
# PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 0.8, 1.6, 1.8, 2.0, 2.2], useAbsoluteValues=True), #so-so
# PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 1.2, 1.8], useAbsoluteValues=True), #|eta| < 0.4, 0.4 < |eta| < 1.2, 1.2 < |eta| < 1.8, |eta| > 1.8,
# PSet(label="TetrajetBjetEta", binLowEdges=[0.2, 0.4, 0.8, 1.2, 1.6, 2.0, 2.2], useAbsoluteValues=True),
# PSet(label="TetrajetBjetEta", binLowEdges=[-1.8, -1.2, -0.4, 0.0, 0.4, 1.2, 1.8], useAbsoluteValues=False),
# PSet(label="TetrajetBjetPt" , binLowEdges=[40, 60, 100, 200, 300], useAbsoluteValues=False), # pT < 40, pT=40-60, pT=60-100, pT=100-200, pT > 200
]
# allSelections.BJetSelection.triggerMatchingApply = True # at least 1 trg b-jet matched to offline b-jets
# allSelections.Trigger.triggerOR = ["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056"]
# allSelections.Trigger.triggerOR = ["HLT_PFHT450_SixJet40_BTagCSV_p056"]
# ================================================================================================
# Command Line Options
# ================================================================================================
# from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import applyAnalysisCommandLineOptions
# applyAnalysisCommandLineOptions(sys.argv, allSelections)
# ================================================================================================
# Build analysis modules
# ================================================================================================
PrintOptions(opts)
builder = AnalysisBuilder(prefix,
dataEras,
searchModes,
usePUreweighting=opts.usePUreweighting,
useTopPtReweighting=opts.useTopPtReweighting,
doSystematicVariations=opts.doSystematics,
analysisType="HToTB",
verbose=opts.verbose)
# Add variations (e.g. for optimisation)
# builder.addVariation("BJetSelection.triggerMatchingApply", [True, False]) # At least 1 trg b-jet dR-matched to offline b-jets
# builder.addVariation("FakeBMeasurement.prelimTopFitChiSqrCutValue", [100, 20])
# builder.addVariation("FakeBMeasurement.prelimTopFitChiSqrCutDirection", ["<=", "==", ">="])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutValue", [0, 1])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutDirection", ["=="])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutDirection", ["<=", "==", ">="])
# builder.addVariation("FakeBMeasurement.numberOfInvertedBJetsCutValue", [0, 1])
# builder.addVariation("FakeBMeasurement.numberOfInvertedBJetsCutDirection", [">="])
# builder.addVariation("FakeBMeasurement.invertedBJetDiscr", "")
# builder.addVariation("FakeBMeasurement.invertedBJetDiscrWorkingPoint", "Loose")
# builder.addVariation("FakeBMeasurement.invertedBJetsSortType", ["Random", "DescendingBDiscriminator"])
# builder.addVariation("FakeBMeasurement.invertedBJetsDiscrMaxCutValue", [0.82, 0.80, 0.75, 0.70])
# builder.addVariation("TopSelection.ChiSqrCutValue", [100])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT450_SixJet40"], ["HLT_PFHT400_SixJet30"]])
# builder.addVariation("TopologySelection.FoxWolframMomentCutValue", [0.5, 0.7])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056"], ["HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056", "HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# Build the builder
builder.build(process, allSelections)
# ================================================================================================
# Example of adding an analyzer whose configuration depends on dataVersion
# ================================================================================================
# def createAnalyzer(dataVersion):
# a = Analyzer("ExampleAnalysis")
# if dataVersion.isMC():
# a.tauPtCut = 10
# else:
# a.tauPtCut = 20
# return a
# process.addAnalyzer("test2", createAnalyzer)
# ================================================================================================
# Pick events
# ================================================================================================
# process.addOptions(EventSaver = PSet(enabled = True,pickEvents = True))
# ================================================================================================
# Run the analysis
# ================================================================================================
# Run the analysis with PROOF? You can give proofWorkers=<N> as a parameter
if opts.jCores:
Print("Running process with PROOF (proofWorkes=%s)" %
(str(opts.jCores)))
process.run(proof=True, proofWorkers=opts.jCores)
else:
Print("Running process (no PROOF)")
process.run()
# Print total time elapsed
tFinish = time.time()
dt = int(tFinish) - int(tStart)
days = divmod(dt, 86400) # days
hours = divmod(days[1], 3600) # hours
mins = divmod(hours[1], 60) # minutes
secs = mins[1] # seconds
Print(
"Total elapsed time is %s days, %s hours, %s mins, %s secs" %
(days[0], hours[0], mins[0], secs), True)
return
from HiggsAnalysis.NtupleAnalysis.main import PSet
import HiggsAnalysis.NtupleAnalysis.parameters.scaleFactors as scaleFactors
##########
## General parameters
##########
##########
## Trigger
##########
trg = PSet(
# No need to specify version numbers, they are automatically scanned in range 1--100 (remove the '_v' suffix)
MuontriggerEfficiencyJsonName="muonPAGEff.json",
# METtriggerEfficiencyJsonName = "metLegTriggerEfficiency_2016_MET90_fit.json",
# L1ETM = 80,
triggerOR=["HLT_Ele27_eta2p1_WPTight_Gsf"],
triggerOR2=[],
)
##########
## MET filter
##########
metFilter = PSet(
discriminators=[ #"hbheNoiseTokenRun2Loose", # Loose is recommended
# "hbheIsoNoiseToken", # under scrutiny
"Flag_HBHENoiseFilter",
"Flag_HBHENoiseIsoFilter",
"Flag_EcalDeadCellTriggerPrimitiveFilter",
# "Flag_CSCTightHaloFilter",
print "Usage: ./QCDMeasurementAnalysis.py <path-to-multicrab-directory> <1pr> <2pr> <3pr>"
sys.exit(0)
from HiggsAnalysis.NtupleAnalysis.main import Process, PSet, Analyzer
from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import obtainAnalysisSuffix
process = Process("QCDMeasurement" + obtainAnalysisSuffix(sys.argv))
process.addDatasetsFromMulticrab(sys.argv[1], blacklist=["ChargedHiggs"])
# Add config
from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import allSelections, applyAnalysisCommandLineOptions, setAngularCutsWorkingPoint
# Enable genuine tau histograms for common plots (needed for calculating N_QCD)
allSelections.CommonPlots.enableGenuineTauHistograms = True
# Set splitting of phase space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
PSet(label="tauPt",
binLowEdges=[60.0, 80.0, 100.0],
useAbsoluteValues=False),
]
#===== Selection customisations
allSelections.TauSelection.prongs = 1
allSelections.TauSelection.tauPtCut = 50.0
allSelections.METSelection.METCutValue = 90.0
allSelections.AngularCutsBackToBack.cutValueJet1 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet2 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet3 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet4 = 40.0
allSelections.TauSelection.rtau = 0.7
allSelections.BJetSelection.bjetDiscrWorkingPoint = "Medium"
#allSelections.AngularCutsCollinear.cutValueJet1 = 80.0
#allSelections.AngularCutsCollinear.cutValueJet2 = 80.0
def _setupToptagEfficiency(topTagPset, topTagEfficiencyFilename, direction,
variationInfo):
'''
Helper function accessed through setupToptagSFInformation
'''
runrange = "runs_273150_284044" #fixme
era = "2016" #fixme
fileName = topTagEfficiencyFilename
fullname = os.path.join(os.getenv("HIGGSANALYSIS_BASE"), "NtupleAnalysis",
"data", fileName)
if not os.path.exists(fullname):
raise Exception(
"Could not find the top-tag efficiency json file! (tried: %s)" %
fullname)
# Read the json file
Print(
"Opening file \"%s\" for reading the top-tag efficiencies" %
(fullname), True)
f = open(fullname)
contents = json.load(f)
f.close()
# Obtain data efficiencies
param = "dataParameters"
if not param in contents.keys():
raise Exception(
"Missing key '%s' in json '%s'! Options: %s" %
(param, fileName, ", ".join(map(str, contents.keys()))))
if not runrange in contents[param].keys():
raise Exception(
"Missing run range '%s' for data in json '%s'! Options: %s" (
runrange, fileName, ", ".join(map(str,
contents[param].keys()))))
datadict = readValues(contents[param][runrange], "data")
# Obtain MC efficiencies
param = "mcParameters"
if not param in contents.keys():
raise Exception(
"Missing key '%s' in json '%s'! Options: %s" %
(param, fileName, ", ".join(map(str, contents.keys()))))
if not era in contents[param].keys():
raise Exception(
"Error: missing era '%s' for mc in json '%s'! Options: %s" (
runrange, fileName, ", ".join(map(str,
contents[param].keys()))))
mcdict = readValues(contents[param][era], "mc")
# Calculate the SF = Eff(Data)/Eff(MC)
keys = datadict.keys()
if len(keys) != len(mcdict.keys()):
raise Exception(
"Different number of bins for data and mc in json '%s'!" %
fileName)
keys.sort()
result = {}
result["binEdges"] = []
result["SF"] = []
result["SFUp"] = []
result["SFDown"] = []
psetList = []
# For-loop: All keys
for i, pT in enumerate(keys, 0):
if i > 0:
result["binEdges"].append(pT)
pTMin = pT
if i == len(keys) - 1:
pTMax = 100000.0 # overflow bin (fixme?)
else:
pTMax = keys[i + 1]
# Get the efficiencies and their errors
effData = datadict[pT]["dataeff"]
effDataUp = datadict[pT]["dataeffup"]
effDataDown = datadict[pT]["dataeffdown"]
effMC = mcdict[pT]["mceff"]
effMCUp = mcdict[pT]["mceffup"]
effMCDown = mcdict[pT]["mceffdown"]
# Define the Scale Factor (SF) as: SF = Eff_Data / Eff_MC
sf = effData / effMC
sfUp = effDataUp / effMC
dsf = (sfUp - sf)
sfDown = sf - dsf # gives symmetric shape
#sfDown = effDataDown / effMC # gives asymmetric shape
Verbose(
"pT = %.1f, sf = %0.3f, sf+ = %0.3f, sf- = %0.3f" %
(pT, sf, sfUp, sfDown), i == 0)
result["SF"].append(sf)
result["SFUp"].append(sfUp)
result["SFDown"].append(sfDown)
if abs(mcdict[pT]["mceffdown"]) < 0.00001:
raise Exception("Down variation in bin '%s' is zero in json '%s'" %
(pT, fileName))
# Sanity check
if result["SF"][len(result["SF"]) - 1] < 0.00001:
raise Exception("In file '%s' bin %s the SF is zero! Please fix!" %
(fileName, pT))
# Define the PSet
p = PSet(
ptMin=pTMin,
ptMax=pTMax,
effMC=effMC,
effMCUp=effMCUp,
effMCDown=effMCDown,
effData=effData,
effDataUp=effDataUp,
effDataDown=effDataDown,
sf=sf,
sfUp=sfUp,
sfDown=sfDown,
)
psetList.append(p)
topTagPset.topTagEfficiency = psetList
if 0:
print topTagPset
#print topTagPset.topTagEfficiency
return
def _setupToptagEffUncertainties(topTagPset, topTagEffUncertaintiesFilename,
direction, variationInfo):
'''
Helper function accessed through setupToptagSFInformation
'''
fullname = os.path.join(os.getenv("HIGGSANALYSIS_BASE"), "NtupleAnalysis",
"data", topTagEffUncertaintiesFilename)
if not os.path.exists(fullname):
raise Exception(
"Could not find the top-tagging eff. uncertainties file! (tried: %s)"
% fullname)
# Read the json file
Print(
"Opening file \"%s\" for reading the top-tag efficiency ucertainties" %
(fullname), True)
f = open(fullname)
contents = json.load(f)
f.close()
# Obtain uncertainties (Use top-mass systematic only once! Take maximum deviation (31 July 2018)
#params = ["TT_hdampUP", "TT_mtop1715", "TT_mtop1755", "TT_fsrdown", "TT_fsrup", "TT_isrdown", "TT_mtop1735", "TT_mtop1785", "TT_TuneEE5C",
# "TT_hdampDOWN", "TT_mtop1695", "TT_evtgen", "TT_ isrup", "TT_mtop1665", "matching"]
params = [
"TT_hdampUP", "TT_fsrdown", "TT_fsrup", "TT_isrdown", "TT_TuneEE5C",
"TT_hdampDOWN", "TT_evtgen", "TT_isrup", "TT_mtop1665", "matching"
]
for param in params:
if not param in contents.keys():
raise Exception(
"Missing key '%s' in json '%s'! Options: %s" %
(param, fullname, ", ".join(map(str, contents.keys()))))
psetList = []
first = contents[params[0]]
firstBins = first["bins"]
# For-loop: All loops
for i in range(0, len(firstBins)):
dSF2 = 0.0
pt = firstBins[i]["pt"]
for param in params:
paramDict = contents[param]
binsList = paramDict["bins"]
case = binsList[i]
pT = case["pt"]
uncertainty = case["uncertainty"]
dSF2 += uncertainty * uncertainty
dSF = math.sqrt(dSF2)
# Find pTMin, pTMax
pTMin = pt
if i == len(firstBins) - 1:
pTMax = 100000.0 # overflow bin (fixme?)
else:
pTMax = firstBins[i + 1]["pt"]
# Uncertainty Up
dSFUp = dSF
# Uncertainty Down
dSFDown = -dSF
# Define a PSet
p = PSet(
ptMin=pTMin,
ptMax=pTMax,
dsfUp=dSFUp,
dsfDown=dSFDown,
)
psetList.append(p)
# Save the PSet
topTagPset.topTagEffUncertainties = psetList
return
btagPayloadFilename="CSVv2.csv",
btagEfficiencyFilename="btageff_HToTB.json",
direction="nominal")
#fakeBTopSelectionBDT = hplus2tbAnalysis.fakeBTopSelectionBDT
fakeBMeasurement = hplus2tbAnalysis.fakeBMeasurement
systTopBDTSelection = PSet(
MuTrijetDRCutValue="2.0",
MuTrijetDRCutDirection=">",
MuBJetDRCutValue="1.5",
MuBJetDRCutDirection="<=",
MiniIsoCutValue="0.1",
MiniIsoCutDirection="<=",
MiniIsoInvCutValue="0.1",
MiniIsoInvCutDirection=">",
METCutValue="50",
METCutDirection=">=",
METInvCutValue="20",
METInvCutDirection="<",
MVACutValue="0.4",
MVACutDirection=">=",
)
#================================================================================================
# Common plots options
#================================================================================================
commonPlotsOptions = hplus2tbAnalysis.commonPlotsOptions
#================================================================================================
# Build all selections group
import HiggsAnalysis.NtupleAnalysis.parameters.scaleFactors as scaleFactors
import HiggsAnalysis.NtupleAnalysis.parameters.jsonReader as jsonReader
#================================================================================================
# General parameters
#================================================================================================
verbose = True
histogramAmbientLevel = "Debug" # (options: "Systematics", "Vital", "Informative", "Debug")
#================================================================================================
# Trigger [scanned in range _v1--_v100 (=>remove the '_v' suffix)]
#================================================================================================
trigger = PSet(
triggerOR=[
"HLT_PFHT400_SixJet30_DoubleBTagCSV_p056",
"HLT_PFHT450_SixJet40_BTagCSV_p056",
"HLT_PFJet450", #for trg eff recovery in 2016H
],
triggerOR2=[],
)
#================================================================================================
# MET filter
#================================================================================================
metFilter = PSet(discriminators=[
"Flag_HBHENoiseFilter", "Flag_HBHENoiseIsoFilter",
"Flag_EcalDeadCellTriggerPrimitiveFilter", "Flag_eeBadScFilter",
"Flag_goodVertices", "Flag_globalTightHalo2016Filter", "badPFMuonFilter",
"badChargedCandidateFilter"
])
#================================================================================================
def main():
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
print __doc__
sys.exit(0)
else:
pass
# ================================================================================================
# Setup the process
# ================================================================================================
maxEvents = {}
maxEvents["All"] = opts.nEvts
# maxEvents["2016"] = 1
# maxEvents["ZZTo4Q"] = -1
# maxEvents["ZJetsToQQ_HT600toInf"] = 1
# maxEvents["WZ_ext1"] = 1
# maxEvents["WZ"] = 1
# maxEvents["WWTo4Q"] = 1
# maxEvents["WJetsToQQ_HT_600ToInf"] = 1
# maxEvents["TTZToQQ"] = 1
# maxEvents["TTWJetsToQQ"] = 1
# maxEvents["TTTT"] = 1
# maxEvents["TT"] = 1
# maxEvents["ST_t_channel_top_4f_inclusiveDecays"] = 1
# maxEvents["ST_t_channel_antitop_4f_inclusiveDecays"] = 1
# maxEvents["ST_tW_top_5f_inclusiveDecays_ext1"] = 1
# maxEvents["ST_tW_top_5f_inclusiveDecays"] = 1
# maxEvents["ST_tW_antitop_5f_inclusiveDecays_ext1"] = 1
# maxEvents["ST_tW_antitop_5f_inclusiveDecays"] = 1
# maxEvents["ST_s_channel_4f_InclusiveDecays"] = 1
# maxEvents["QCD_HT700to1000_ext1"] = 1
# maxEvents["QCD_HT700to1000"] = 1
# maxEvents["QCD_HT50to100"] = 1
# maxEvents["QCD_HT500to700_ext1"] = 1
# maxEvents["QCD_HT500to700"] = 1
# maxEvents["QCD_HT300to500_ext1"] = 1
# maxEvents["QCD_HT300to500"] = 1
# maxEvents["QCD_HT200to300_ext1"] = 1
# maxEvents["QCD_HT200to300"] = 1
# maxEvents["QCD_HT2000toInf_ext1"] = 1
# maxEvents["QCD_HT2000toInf"] = 1
# maxEvents["QCD_HT1500to2000_ext1"] = 1
# maxEvents["QCD_HT1500to2000"] = 1
# maxEvents["QCD_HT100to200"] = 1
# maxEvents["QCD_HT1000to1500_ext1"] = 1
# maxEvents["QCD_HT1000to1500"] = 1
# maxEvents["JetHT_Run2016H_03Feb2017_ver3_v1_284036_284044"] = 1
# maxEvents["JetHT_Run2016H_03Feb2017_ver2_v1_281613_284035"] = 1
# maxEvents["JetHT_Run2016G_03Feb2017_v1_278820_280385"] = 1
# maxEvents["JetHT_Run2016F_03Feb2017_v1_278801_278808"] = 1
# maxEvents["JetHT_Run2016F_03Feb2017_v1_277932_278800"] = 1
# maxEvents["JetHT_Run2016E_03Feb2017_v1_276831_277420"] = 1
# maxEvents["JetHT_Run2016D_03Feb2017_v1_276315_276811"] = 1
# maxEvents["JetHT_Run2016C_03Feb2017_v1_275656_276283"] = 1
# maxEvents["JetHT_Run2016B_03Feb2017_ver2_v2_273150_275376"] = 1
# maxEvents["DYJetsToQQ_HT180"] = 1
# maxEvents["ChargedHiggs_HplusTB_HplusToTB_M_500"] = 1
process = Process(prefix, postfix, maxEvents)
# ================================================================================================
# Add the datasets (according to user options)
# ================================================================================================
if (opts.includeOnlyTasks):
Print("Adding only dataset %s from multiCRAB directory %s" %
(opts.includeOnlyTasks, opts.mcrab))
process.addDatasetsFromMulticrab(
opts.mcrab, includeOnlyTasks=opts.includeOnlyTasks)
elif (opts.excludeTasks):
Print("Adding all datasets except %s from multiCRAB directory %s" %
(opts.excludeTasks, opts.mcrab))
Print(
"If collision data are present, then vertex reweighting is done according to the chosen data era (era=2015C, 2015D, 2015) etc..."
)
process.addDatasetsFromMulticrab(opts.mcrab,
excludeTasks=opts.excludeTasks)
else:
myBlackList = [
"M_180", "M_2500", "M_3000", "M_5000", "M_7000", "M_10000", "QCD"
]
Print("Adding all datasets from multiCRAB directory %s except %s" %
(opts.mcrab, (",".join(myBlackList))))
Print(
"Vertex reweighting is done according to the chosen data era (%s)"
% (",".join(dataEras)))
# process.addDatasetsFromMulticrab(opts.mcrab, blacklist=myBlackList)
if len(myBlackList) > 0:
regex = "|".join(myBlackList)
process.addDatasetsFromMulticrab(opts.mcrab, excludeTasks=regex)
else:
process.addDatasetsFromMulticrab(opts.mcrab)
# ================================================================================================
# Overwrite Default Settings
# ================================================================================================
from HiggsAnalysis.NtupleAnalysis.parameters.hplus2tbAnalysis import allSelections
allSelections.verbose = opts.verbose
allSelections.histogramAmbientLevel = opts.histoLevel
# Set splitting of phase-space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
PSet(label="TetrajetBjetEta",
binLowEdges=[0.4, 1.2, 1.8],
useAbsoluteValues=True),
#PSet(label="TetrajetBjetEta", binLowEdges=[0.4, 1.2], useAbsoluteValues=True),
#PSet(label="TetrajetBjetEta", binLowEdges=[1.2], useAbsoluteValues=True),
#PSet(label="TetrajetBjetEta", binLowEdges=[1.3], useAbsoluteValues=True), # good but perhaps limit to 1.2
#PSet(label="TetrajetBjetEta", binLowEdges=[0.6, 1.4], useAbsoluteValues=True), #middle bin is bad
]
# allSelections.BJetSelection.triggerMatchingApply = True # at least 1 trg b-jet matched to offline b-jets
# allSelections.Trigger.triggerOR = ["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056"]
# allSelections.Trigger.triggerOR = ["HLT_PFHT450_SixJet40_BTagCSV_p056"]
# ================================================================================================
# Command Line Options
# ================================================================================================
# from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import applyAnalysisCommandLineOptions
# applyAnalysisCommandLineOptions(sys.argv, allSelections)
# ================================================================================================
# Build analysis modules
# ================================================================================================
PrintOptions(opts)
builder = AnalysisBuilder(prefix,
dataEras,
searchModes,
usePUreweighting=opts.usePUreweighting,
useTopPtReweighting=opts.useTopPtReweighting,
doSystematicVariations=opts.doSystematics)
# Add variations (e.g. for optimisation)
# builder.addVariation("BJetSelection.triggerMatchingApply", [True, False]) # At least 1 trg b-jet dR-matched to offline b-jets
# builder.addVariation("TestQGLR.prelimTopFitChiSqrCutValue", [100, 20])
# builder.addVariation("TestQGLR.prelimTopFitChiSqrCutDirection", ["<=", "==", ">="])
# builder.addVariation("TestQGLR.numberOfBJetsCutValue", [0, 1])
# builder.addVariation("TestQGLR.numberOfBJetsCutDirection", ["=="])
# builder.addVariation("TestQGLR.numberOfBJetsCutDirection", ["<=", "==", ">="])
# builder.addVariation("TestQGLR.numberOfInvertedBJetsCutValue", [0, 1])
# builder.addVariation("TestQGLR.numberOfInvertedBJetsCutDirection", [">="])
# builder.addVariation("TestQGLR.invertedBJetDiscr", "")
# builder.addVariation("TestQGLR.invertedBJetDiscrWorkingPoint", "Loose")
# builder.addVariation("TestQGLR.invertedBJetsSortType", ["Random", "DescendingBDiscriminator"])
# builder.addVariation("TestQGLR.invertedBJetsDiscrMaxCutValue", [0.82, 0.80, 0.75, 0.70])
# builder.addVariation("TopSelection.ChiSqrCutValue", [100])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT450_SixJet40"], ["HLT_PFHT400_SixJet30"]])
# builder.addVariation("TopologySelection.FoxWolframMomentCutValue", [0.5, 0.7])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056"], ["HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056", "HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# Build the builder
builder.build(process, allSelections)
# ================================================================================================
# Example of adding an analyzer whose configuration depends on dataVersion
# ================================================================================================
# def createAnalyzer(dataVersion):
# a = Analyzer("ExampleAnalysis")
# if dataVersion.isMC():
# a.tauPtCut = 10
# else:
# a.tauPtCut = 20
# return a
# process.addAnalyzer("test2", createAnalyzer)
# ================================================================================================
# Pick events
# ================================================================================================
# process.addOptions(EventSaver = PSet(enabled = True,pickEvents = True))
# ================================================================================================
# Run the analysis
# ================================================================================================
# Run the analysis with PROOF? You can give proofWorkers=<N> as a parameter
if opts.jCores:
Print("Running process with PROOF (proofWorkes=%s)" %
(str(opts.jCores)))
process.run(proof=True, proofWorkers=opts.jCores)
else:
Print("Running process (no PROOF)")
process.run()
return
def main():
# Save start time (epoch seconds)
tStart = time.time()
Verbose("Started @ " + str(tStart), True)
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
print __doc__
sys.exit(0)
else:
pass
# ================================================================================================
# Setup the process
# ================================================================================================
maxEvents = {}
maxEvents["All"] = opts.nEvts
process = Process(prefix, postfix, maxEvents)
# ================================================================================================
# Add the datasets (according to user options)
# ================================================================================================
if (opts.includeOnlyTasks):
Print("Adding only dataset %s from multiCRAB directory %s" %
(opts.includeOnlyTasks, opts.mcrab))
process.addDatasetsFromMulticrab(
opts.mcrab, includeOnlyTasks=opts.includeOnlyTasks)
elif (opts.excludeTasks):
Print("Adding all datasets except %s from multiCRAB directory %s" %
(opts.excludeTasks, opts.mcrab))
Print(
"If collision data are present, then vertex reweighting is done according to the chosen data era (era=2015C, 2015D, 2015) etc..."
)
process.addDatasetsFromMulticrab(opts.mcrab,
excludeTasks=opts.excludeTasks)
else:
myBlackList = [
"QCD_b"
"ChargedHiggs_HplusTB_HplusToTB_M_180_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_200_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_220_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_250_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_300_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_350_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_400_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_500_ext1",
#"ChargedHiggs_HplusTB_HplusToTB_M_650", #10M events!
"ChargedHiggs_HplusTB_HplusToTB_M_800_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_1000_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_2000_ext1"
"ChargedHiggs_HplusTB_HplusToTB_M_2500_ext1",
"ChargedHiggs_HplusTB_HplusToTB_M_3000_ext1",
# "ChargedHiggs_HplusTB_HplusToTB_M_1000",
"ChargedHiggs_HplusTB_HplusToTB_M_1500", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_2000", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_2500", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_3000", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_5000", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_7000", # Speeed things up
"ChargedHiggs_HplusTB_HplusToTB_M_10000", # Speeed things up
]
if opts.doSystematics:
myBlackList.append("QCD")
Print("Adding all datasets from multiCRAB directory %s" % (opts.mcrab))
Print(
"If collision data are present, then vertex reweighting is done according to the chosen data era (era=2015C, 2015D, 2015) etc..."
)
regex = "|".join(myBlackList)
if len(myBlackList) > 0:
process.addDatasetsFromMulticrab(opts.mcrab, excludeTasks=regex)
else:
process.addDatasetsFromMulticrab(opts.mcrab)
# ================================================================================================
# Overwrite Default Settings
# ================================================================================================
from HiggsAnalysis.NtupleAnalysis.parameters.jetTriggers import allSelections
from HiggsAnalysis.NtupleAnalysis.main import PSet
import HiggsAnalysis.NtupleAnalysis.parameters.scaleFactors as scaleFactors
allSelections.verbose = opts.verbose
allSelections.histogramAmbientLevel = opts.histoLevel
#==========================
# Systematics selections
#==========================
# marina
allSelections.SystTopBDTSelection.MVACutValue = 0.40
allSelections.TopSelectionBDT.TopMVACutValue = 0.40
# BDT MisID SF
MisIDSF = PSet(
MisIDSFJsonName="topMisID_BDT0p40_TopMassCut400.json",
ApplyMisIDSF=False,
)
scaleFactors.assignMisIDSF(MisIDSF, "nominal", MisIDSF.MisIDSFJsonName)
allSelections.MisIDSF = MisIDSF
allSelections.SystTopBDTSelection.MiniIsoCutValue = 0.1
allSelections.SystTopBDTSelection.MiniIsoInvCutValue = 0.1
allSelections.SystTopBDTSelection.METCutValue = 50.0
allSelections.SystTopBDTSelection.METInvCutValue = 20.0
# Muon
allSelections.MuonSelection.muonPtCut = 30
# Jets
allSelections.JetSelection.numberOfJetsCutValue = 4
allSelections.JetSelection.jetPtCuts = [40.0, 40.0, 40.0, 40.0]
# Trigger
allSelections.Trigger.triggerOR = ["HLT_Mu50"]
# Bjets
allSelections.BJetSelection.jetPtCuts = [40.0, 40.0]
allSelections.BJetSelection.numberOfBJetsCutValue = 2
# ================================================================================================
# Add Analysis Variations
# ================================================================================================
# selections = allSelections.clone()
# process.addAnalyzer(prefix, Analyzer(prefix, config=selections, silent=False) ) #trigger passed from selections
# ================================================================================================
# Command Line Options
# ================================================================================================
# from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import applyAnalysisCommandLineOptions
# applyAnalysisCommandLineOptions(sys.argv, allSelections)
#================================================================================================
# Build analysis modules
#================================================================================================
PrintOptions(opts)
builder = AnalysisBuilder(prefix,
dataEras,
searchModes,
usePUreweighting=opts.usePUreweighting,
useTopPtReweighting=opts.useTopPtReweighting,
doSystematicVariations=opts.doSystematics,
analysisType="HToTB")
# Add variations (e.g. for optimisation)
# builder.addVariation("METSelection.METCutValue", [100,120,140])
# builder.addVariation("AngularCutsBackToBack.workingPoint", ["Loose","Medium","Tight"])
# builder.addVariation("BJetSelection.triggerMatchingApply", [False])
# builder.addVariation("TopSelection.ChiSqrCutValue", [5, 10, 15, 20])
# Build the builder
builder.build(process, allSelections)
# ================================================================================================
# Example of adding an analyzer whose configuration depends on dataVersion
# ================================================================================================
#def createAnalyzer(dataVersion):
#a = Analyzer("ExampleAnalysis")
#if dataVersion.isMC():
#a.tauPtCut = 10
#else:
#a.tauPtCut = 20
#return a
#process.addAnalyzer("test2", createAnalyzer)
# ================================================================================================
# Pick events
# ================================================================================================
#process.addOptions(EventSaver = PSet(enabled = True,pickEvents = True))
# ================================================================================================
# Run the analysis
# ================================================================================================
# Run the analysis with PROOF? You can give proofWorkers=<N> as a parameter
if opts.jCores:
Print("Running process with PROOF (proofWorkes=%s)" %
(str(opts.jCores)))
process.run(proof=True, proofWorkers=opts.jCores)
else:
Print("Running process")
process.run()
# Print total time elapsed
tFinish = time.time()
dt = int(tFinish) - int(tStart)
days = divmod(dt, 86400) # days
hours = divmod(days[1], 3600) # hours
mins = divmod(hours[1], 60) # minutes
secs = mins[1] # seconds
Print(
"Total elapsed time is %s days, %s hours, %s mins, %s secs" %
(days[0], hours[0], mins[0], secs), True)
return
import HiggsAnalysis.NtupleAnalysis.parameters.scaleFactors as scaleFactors
#================================================================================================
# General parameters
#================================================================================================
verbose = True
histogramAmbientLevel = "Debug" # Options: Systematics, Vital, Informative, Debug
#================================================================================================
# Trigger
#================================================================================================
trigger = PSet(
triggerOR=[
"HLT_PFHT400_SixJet30_DoubleBTagCSV_p056", # scanned in range _v1--_v100 (=>remove the '_v' suffix)
"HLT_PFHT450_SixJet40_BTagCSV_p056", # scanned in range _v1--_v100 (=>remove the '_v' suffix)
#"HLT_PFHT400_SixJet30", #Prescale 110 at inst. lumi 1.35E+34
#"HLT_PFHT450_SixJet40", #Prescale 26 at inst. lumi 1.35E+34
],
triggerOR2=[],
)
#================================================================================================
# Tau selection (sync with HToTauNu analysis)
#================================================================================================
tauSelection = PSet(
applyTriggerMatching=False,
triggerMatchingCone=0.1, # DeltaR for matching offline tau with trigger tau
tauPtCut=20.0, #
tauEtaCut=2.3, #
tauLdgTrkPtCut=0.0, #
prongs=-1, # options: 1, 2, 3, 12, 13, 23, 123 or -1 (all)
def createAnalyzer(dataVersion, era):
a = Analyzer(
"TriggerEfficiency",
name=era,
Trigger=PSet(triggerOR=[], triggerOR2=[]),
METFilter=metFilter,
usePileupWeights=True,
offlineSelection=leg,
MuonSelection=PSet(
# discriminators = ["muIDMedium"],
# discriminators = ["TrgMatch_IsoMu20_eta2p1"],
discriminators=["Muons_TrgMatch_IsoMu16_eta2p1"], ),
TauSelection=PSet(
discriminators=[
"byLooseCombinedIsolationDeltaBetaCorr3Hits", #"byMediumIsolationMVA3newDMwLT",
"againstMuonTight3",
"againstElectronMediumMVA6"
],
nprongs=1,
relaxedOfflineSelection=False),
binning=binning,
xLabel=xLabel,
yLabel=yLabel,
)
if isData(dataVersion):
a.Trigger.triggerOR = [
"HLT_IsoMu15_eta2p1_L1ETM20_v3", "HLT_IsoMu15_eta2p1_L1ETM20_v4",
"HLT_IsoMu15_eta2p1_L1ETM20_v5", "HLT_IsoMu15_eta2p1_L1ETM20_v6",
"HLT_IsoMu15_eta2p1_L1ETM20_v7"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v2",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v4",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v6",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v7",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v9",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v10"
]
if era == "2015C":
# a.Trigger.triggerOR = ["HLT_IsoMu16_eta2p1_CaloMET30_v1",
# "HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu16_eta2p1_CaloMET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
# "HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1"]
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v2"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v2"]
if era == "2015D":
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v3",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx",
"HLT_IsoMu16_eta2p1_MET30_v1", "HLT_IsoMu16_eta2p1_MET30_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v3",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
if era == "2015CD":
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v3",
"HLT_IsoMu16_eta2p1_MET30_v1"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v3",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_v1"
]
if "2016" in era:
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_vx", "HLT_IsoMu22_eta2p1_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx",
"HLT_IsoMu21_eta2p1_LooseIsoPFTau50_Trk30_eta2p1_SingleL1_vx"
]
if era == "2016ICHEP":
a.Trigger.triggerOR = ["HLT_IsoMu16_eta2p1_MET30_vx"]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
if era == "2016HIPFIXED":
a.Trigger.triggerOR = ["HLT_IsoMu22_eta2p1_vx"]
a.Trigger.triggerOR2 = [
"HLT_IsoMu21_eta2p1_LooseIsoPFTau50_Trk30_eta2p1_SingleL1_vx"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v1",
# "HLT_IsoMu20_eta2p1_v2",
# "HLT_IsoMu17_eta2p1_v2"]
# a.Trigger.triggerOR2= ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v1",
# "HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v2"]
# runmin,runmax = process.getRuns()
# a.lumi = lumi
a.runMin = runmin
a.runMax = runmax
else:
a.Trigger.triggerOR = ["HLT_IsoMu15_eta2p1_L1ETM20_v5"]
a.Trigger.triggerOR2 = [
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v6"
]
if era == "2015C" or era == "2015D" or era == "2015CD" or "2016" in era:
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_CaloMET30_v1",
"HLT_IsoMu16_eta2p1_MET30_vx",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_CaloMET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v1"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v1"]
#print "check triggerOR",a.Trigger.triggerOR
return a
import HiggsAnalysis.NtupleAnalysis.parameters.jsonReader as jsonReader
#================================================================================================
# General parameters
#================================================================================================
verbose = True
histogramAmbientLevel = "Debug" # Options: Systematics, Vital, Informative, Debug
#================================================================================================
# Trigger
#================================================================================================
trigger = PSet(
# scanned in range _v1--_v100 (=>remove the '_v' suffix)
triggerOR=[
"HLT_PFHT400_SixJet30_DoubleBTagCSV_p056",
"HLT_PFHT450_SixJet40_BTagCSV_p056",
"HLT_PFJet450", #for trg eff recovery in 2016H
],
triggerOR2=[],
)
#================================================================================================
# MET filter
#================================================================================================
metFilter = PSet(discriminators=[
"hbheNoiseTokenRun2Loose", "Flag_HBHENoiseIsoFilter",
"Flag_EcalDeadCellTriggerPrimitiveFilter", "Flag_CSCTightHaloFilter",
"Flag_eeBadScFilter", "Flag_goodVertices"
])
#================================================================================================
#!/usr/bin/env python
from HiggsAnalysis.NtupleAnalysis.main import PSet
import HiggsAnalysis.NtupleAnalysis.parameters.scaleFactors as scaleFactors
#====== General parameters
histoLevel = "Debug" # Options: Systematics, Vital, Informative, Debug
#====== Trigger
trg = PSet(
# No need to specify version numbers, they are automatically scanned in range 1--100 (remove the '_v' suffix)
L1ETM = 80,
triggerOR = ["HLT_LooseIsoPFTau50_Trk30_eta2p1_MET90"
],
triggerOR2 = [],
)
#====== MET filter
metFilter = PSet(
discriminators = [#"hbheNoiseTokenRun2Loose", # Loose is recommended
# "hbheIsoNoiseToken", # under scrutiny
"Flag_HBHENoiseFilter",
"Flag_HBHENoiseIsoFilter",
"Flag_EcalDeadCellTriggerPrimitiveFilter",
# "Flag_CSCTightHaloFilter",
"Flag_eeBadScFilter",
"Flag_goodVertices",
"Flag_globalTightHalo2016Filter",
"badPFMuonFilter",
"badChargedCandidateFilter"]
def createAnalyzer(dataVersion, era):
a = Analyzer(
"TriggerEfficiency",
name=era,
Trigger=PSet(triggerOR=[], triggerOR2=[]),
usePileupWeights=True,
offlineSelection=leg,
MuonSelection=PSet(
# discriminators = ["muIDMedium"],
# discriminators = ["TrgMatch_IsoMu20_eta2p1"],
discriminators=["Muons_TrgMatch_IsoMu16_eta2p1"], ),
TauSelection=PSet(discriminators=[
"byLooseCombinedIsolationDeltaBetaCorr3Hits", "againstMuonTight3",
"againstElectronMediumMVA5"
], ),
binning=binning,
xLabel=xLabel,
yLabel=yLabel,
)
if isData(dataVersion):
a.Trigger.triggerOR = [
"HLT_IsoMu15_eta2p1_L1ETM20_v3", "HLT_IsoMu15_eta2p1_L1ETM20_v4",
"HLT_IsoMu15_eta2p1_L1ETM20_v5", "HLT_IsoMu15_eta2p1_L1ETM20_v6",
"HLT_IsoMu15_eta2p1_L1ETM20_v7"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v2",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v4",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v6",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v7",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v9",
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v10"
]
if era == "2015C":
# a.Trigger.triggerOR = ["HLT_IsoMu16_eta2p1_CaloMET30_v1",
# "HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu16_eta2p1_CaloMET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
# "HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1"]
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v2"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v2"]
if era == "2015D":
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v3",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx",
"HLT_IsoMu16_eta2p1_MET30_v1", "HLT_IsoMu16_eta2p1_MET30_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v3",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
if era == "2015CD":
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_v3",
"HLT_IsoMu16_eta2p1_MET30_v1"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v2",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_v3",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_v1"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v1",
# "HLT_IsoMu20_eta2p1_v2",
# "HLT_IsoMu17_eta2p1_v2"]
# a.Trigger.triggerOR2= ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v1",
# "HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v2"]
lumi, runmin, runmax = runRange(era)
a.lumi = lumi
a.runMin = runmin
a.runMax = runmax
else:
a.Trigger.triggerOR = ["HLT_IsoMu15_eta2p1_L1ETM20_v5"]
a.Trigger.triggerOR2 = [
"HLT_IsoMu15_eta2p1_LooseIsoPFTau35_Trk20_Prong1_L1ETM20_v6"
]
if era == "2015C" or era == "2015D" or era == "2015CD":
a.Trigger.triggerOR = [
"HLT_IsoMu16_eta2p1_CaloMET30_v1",
"HLT_IsoMu16_eta2p1_MET30_vx",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_vx"
]
a.Trigger.triggerOR2 = [
"HLT_IsoMu16_eta2p1_CaloMET30_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_IsoMu16_eta2p1_MET30_LooseIsoPFTau50_Trk30_eta2p1_vx",
"HLT_IsoMu16_eta2p1_MET30_JetIdCleaned_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
# a.Trigger.triggerOR = ["HLT_IsoMu20_eta2p1_v1"]
# a.Trigger.triggerOR2 = ["HLT_IsoMu17_eta2p1_LooseIsoPFTau20_v1"]
#print "check triggerOR",a.Trigger.triggerOR
return a
def _setupBtagSFDatabase(btagPset, btagPayloadFilename, direction,
variationInfo):
fullname = os.path.join(os.getenv("HIGGSANALYSIS_BASE"), "NtupleAnalysis",
"data", btagPayloadFilename)
if not os.path.exists(fullname):
raise Exception(
"Error: Could not find btag POG btag SF payload csv file! (tried: %s)"
% fullname)
# Obtain header row and rows
headerRow = None
rows = []
validAlgoHeaderPairs = {
"pfCombinedInclusiveSecondaryVertexV2BJetTags": "CSVv2"
}
if not btagPset.__getattr__("bjetDiscr") in validAlgoHeaderPairs.keys():
raise Exception(
"Error: No valid payload header ID has been specified for btag algo %s"
% btagPset.__getattr__("bjetDiscr"))
directionLUT = {"nominal": " central", "down": " down", "up": " up"}
if not direction in directionLUT.keys():
raise Exception("Error: direction '%s' is unknown! Valid options: %s" %
(direction, ", ".join(map(str, directionLUT))))
workingPointLUT = {
"Loose": "0",
"Medium": "1",
"Tight": "2",
}
if not btagPset.__getattr__(
"bjetDiscrWorkingPoint") in workingPointLUT.keys():
raise Exception(
"Error: Btag working point '%s' is not defined in the look-up table!"
% (btagPset.__getattr__("bjetDiscrWorkingPoint")))
# Column names in the btag payload file
headerColumnIndices = {
"OperatingPoint": None,
"measurementType": None,
"sysType": None,
"jetFlavor": None,
"etaMin": None,
"etaMax": None,
"ptMin": None,
"ptMax": None,
"discrMin": None,
"discrMax": None,
"formula": None
}
with open(fullname, 'rb') as csvfile:
reader = csv.reader(csvfile)
for row in reader:
if headerRow == None:
headerRow = row[1:]
# Check that payload matches with tagger
if validAlgoHeaderPairs[btagPset.__getattr__(
"bjetDiscr")] != row[0]:
raise Exception(
"Error: btag algo = %s is incompatible with btag SF payload file header '%s' (expected %s)!"
% (btagPset.__getattr__("bjetDiscr"), row[0],
validAlgoHeaderPairs[btagPset.__getattr__(
"bjetDiscr")]))
# Obtain column indices
for key in headerColumnIndices.keys():
for i in range(len(headerRow)):
if headerRow[i] == key or headerRow[
i] == " " + key or headerRow[
i] == " " + key + " ":
headerColumnIndices[key] = i
if headerColumnIndices[key] == None:
raise Exception(
"Error: could not find column '%s' in file %s:\n header = %s"
% (key, fullname, headerRow))
else:
# Store only the rows which apply for the desired variation and working point
if row[headerColumnIndices[
"OperatingPoint"]] == workingPointLUT[
btagPset.__getattr__("bjetDiscrWorkingPoint")]:
rows.append(row)
if len(rows) == 0:
raise Exception(
"Error: for unknown reason, no entries found from the btag SF payload (%s)!"
% fullname)
# Convert output into vector of PSets
psetList = []
for row in rows:
p = PSet(jetFlavor=int(row[headerColumnIndices["jetFlavor"]]),
ptMin=float(row[headerColumnIndices["ptMin"]]),
ptMax=float(row[headerColumnIndices["ptMax"]]),
etaMin=float(row[headerColumnIndices["etaMin"]]),
etaMax=float(row[headerColumnIndices["etaMax"]]),
discrMin=float(row[headerColumnIndices["discrMin"]]),
discrMax=float(row[headerColumnIndices["discrMax"]]),
sysType=row[headerColumnIndices["sysType"]],
formula=row[headerColumnIndices["formula"]])
psetList.append(p)
btagPset.btagSF = psetList
print "Usage: ./QCDMeasurementAnalysis.py <path-to-multicrab-directory> <1pr> <2pr> <3pr>"
sys.exit(0)
from HiggsAnalysis.NtupleAnalysis.main import Process, PSet, Analyzer
from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import obtainAnalysisSuffix
process = Process("QCDMeasurement" + obtainAnalysisSuffix(sys.argv))
process.addDatasetsFromMulticrab(sys.argv[1], blacklist=["ChargedHiggs"])
# Add config
from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import allSelections, applyAnalysisCommandLineOptions, setAngularCutsWorkingPoint
# Enable genuine tau histograms for common plots (needed for calculating N_QCD)
allSelections.CommonPlots.enableGenuineTauHistograms = True
# Set splitting of phase space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
PSet(label="tauPt",
binLowEdges=[60.0, 80.0, 100.0],
useAbsoluteValues=False),
PSet(label="tauEta", binLowEdges=[0.6, 1.4], useAbsoluteValues=True),
]
#===== Selection customisations
allSelections.TauSelection.prongs = 1
allSelections.TauSelection.tauPtCut = 50.0
allSelections.METSelection.METCutValue = 90.0
allSelections.AngularCutsBackToBack.cutValueJet1 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet2 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet3 = 40.0
allSelections.AngularCutsBackToBack.cutValueJet4 = 40.0
allSelections.TauSelection.rtau = 0.75
#allSelections.TauSelection.rtauSmallerThan = 0.75
allSelections.BJetSelection.bjetDiscrWorkingPoint = "Medium"
def createAnalyzer(dataVersion, era, onlineSelection="MET80"):
useCaloMET = False
if "CaloMET" in era:
useCaloMET = True
era = era[:-8]
a = Analyzer(
"TriggerEfficiency",
name=era,
Trigger=PSet(triggerOR=[], triggerOR2=[]),
usePileupWeights=True,
# usePileupWeights = False,
onlineSelection=onlineSelection,
offlineSelection=leg,
TauSelection=signalAnalysis.tauSelection,
# TauSelection = PSet(
# discriminators = ["byLooseCombinedIsolationDeltaBetaCorr3Hits",
# "againstMuonTight3",
# "againstElectronMediumMVA5"],
# ),
ElectronSelection=signalAnalysis.eVeto,
MuonSelection=signalAnalysis.muVeto,
JetSelection=signalAnalysis.jetSelection,
BJetSelection=signalAnalysis.bjetSelection,
binning=binning,
xLabel=xLabel,
yLabel=yLabel,
)
# a.TauSelection.applyTriggerMatching = False
a.JetSelection.numberOfJetsCutValue = 3
# a.BJetSelection.bjetDiscrWorkingPoint = "Medium"
a.BJetSelection.numberOfBJetsCutValue = 1
if dataVersion.isData():
a.Trigger.triggerOR = [
"HLT_LooseIsoPFTau35_Trk20_Prong1_v2",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v3",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v4",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v6",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v7",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v9",
"HLT_LooseIsoPFTau35_Trk20_Prong1_v10"
]
a.Trigger.triggerOR2 = [
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v2",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v3",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v4",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v6",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v7",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v9",
"HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v10"
]
if era == "2015C" or era == "2015D" or era == "2015CD":
a.Trigger.triggerOR = [
"HLT_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_v2",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_v3",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
a.Trigger.triggerOR2 = [
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection +
"_JetIdCleaned_v1", "HLT_LooseIsoPFTau50_Trk30_eta2p1_" +
onlineSelection + "_JetIdCleaned_v2",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection + "_v1",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection +
"_JetIdCleaned_vx",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection + "_vx"
]
lumi, runmin, runmax = runRange(era)
a.lumi = lumi
a.runMin = runmin
a.runMax = runmax
else:
a.Trigger.triggerOR = ["HLT_LooseIsoPFTau35_Trk20_Prong1_v6"]
a.Trigger.triggerOR2 = ["HLT_LooseIsoPFTau35_Trk20_Prong1_MET70_v6"]
if era == "2015C" or era == "2015D" or era == "2015CD":
a.Trigger.triggerOR = [
"HLT_LooseIsoPFTau50_Trk30_eta2p1_v1",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_vx"
]
a.Trigger.triggerOR2 = [
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection + "_v1",
"HLT_LooseIsoPFTau50_Trk30_eta2p1_" + onlineSelection + "_vx"
]
if useCaloMET:
a.Trigger.triggerOR2 = []
return a
def main():
# Save start time (epoch seconds)
tStart = time.time()
Verbose("Started @ " + str(tStart), True)
# Require at least two arguments (script-name, path to multicrab)
if len(sys.argv) < 2:
Print(
"Not enough arguments passed to script execution. Printing docstring & EXIT."
)
print __doc__
sys.exit(0)
else:
pass
# ================================================================================================
# Setup the process
# ================================================================================================
completeList = GetDatasetCompleteList()
whiteList = GetDatasetWhitelist(opts)
blackList = GetDatasetBlackList(completeList, whiteList)
maxEvents = {}
for d in whiteList:
maxEvents[d] = -1
#if d == "ChargedHiggs_HplusTB_HplusToTB_M_650":
# maxEvents[d] = 4000000
process = Process(prefix, postfix, maxEvents)
# ================================================================================================
# Add the datasets (according to user options)
# ================================================================================================
if (opts.includeOnlyTasks):
Verbose("Adding only dataset %s from multiCRAB directory %s" %
(opts.includeOnlyTasks, opts.mcrab))
process.addDatasetsFromMulticrab(
opts.mcrab, includeOnlyTasks=opts.includeOnlyTasks)
elif (opts.excludeTasks):
Verbose("Adding all datasets except %s from multiCRAB directory %s" %
(opts.excludeTasks, opts.mcrab))
Print(
"If collision data are present, then vertex reweighting is done according to the chosen data era (era=2015C, 2015D, 2015) etc..."
)
process.addDatasetsFromMulticrab(opts.mcrab,
excludeTasks=opts.excludeTasks)
else:
myBlackList = blackList
#myBlackList = ["M_180", "M_200" , "M_220" , "M_250" , "M_300" , "M_350" , "M_400" , "M_500" , "M_650",
# "M_800", "M_1000", "M_1500", "M_2000", "M_2500", "M_3000", "M_5000", "M_7000", "M_10000"]
# Extend the blacklist with datasets not in the group
#myBlackList.extend(blackList)
Verbose("Adding all datasets from multiCRAB directory %s except %s" %
(opts.mcrab, (",".join(myBlackList))))
Verbose(
"Vertex reweighting is done according to the chosen data era (%s)"
% (",".join(dataEras)))
regex = "|".join(myBlackList)
if len(myBlackList) > 0:
process.addDatasetsFromMulticrab(opts.mcrab, excludeTasks=regex)
else:
process.addDatasetsFromMulticrab(opts.mcrab)
# ================================================================================================
# Overwrite Default Settings
# ================================================================================================
from HiggsAnalysis.NtupleAnalysis.parameters.hplus2tbAnalysis import allSelections
allSelections.verbose = opts.verbose
allSelections.histogramAmbientLevel = opts.histoLevel
# Set splitting of phase-space (first bin is below first edge value and last bin is above last edge value)
allSelections.CommonPlots.histogramSplitting = [
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 160, 300], useAbsoluteValues=False), # last (not bad)
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 120, 300], useAbsoluteValues=False), # last (not bad)
#PSet(label="TetrajetBjetPt" , binLowEdges=[50, 150, 300], useAbsoluteValues=False), # last (quite good)
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 60, 150], useAbsoluteValues=False),
# PSet(label="TetrajetBjetPt" , binLowEdges=[60, 90, 160, 300], useAbsoluteValues=False), # BEST (v1)
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 60, 90, 160, 330], useAbsoluteValues=False),
PSet(label="TetrajetBjetPt",
binLowEdges=[60, 90, 160, 300],
useAbsoluteValues=False), # BEST (v2)
#PSet(label="TetrajetBjetPt" , binLowEdges=[60, 90, 160, 300], useAbsoluteValues=False), # BEST (v3)
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 65, 130, 330], useAbsoluteValues=False), # BEST (v4)
# PSet(label="TetrajetBjetPt" , binLowEdges=[50, 65, 95, 130, 330], useAbsoluteValues=False), # BEST (v5)
# PSet(label="TetrajetBjetPt" , binLowEdges=[45, 50, 60, 90, 160, 330], useAbsoluteValues=False), # BEST (v5)
#
PSet(label="TetrajetBjetEta",
binLowEdges=[0.8, 1.4, 2.0],
useAbsoluteValues=True), # optimised
]
# Overwrite values
# allSelections.TopSelectionBDT.CSV_bDiscCutDirection = ">="
allSelections.TopSelectionBDT.CSV_bDiscCutValue = 0.54 # allow CSVv2-L for inverted top
# ================================================================================================
# Command Line Options
# ================================================================================================
# from HiggsAnalysis.NtupleAnalysis.parameters.signalAnalysisParameters import applyAnalysisCommandLineOptions
# applyAnalysisCommandLineOptions(sys.argv, allSelections)
# ================================================================================================
# Build analysis modules
# ================================================================================================
PrintOptions(opts)
builder = AnalysisBuilder(prefix,
dataEras,
searchModes,
usePUreweighting=opts.usePUreweighting,
useTopPtReweighting=opts.useTopPtReweighting,
doSystematicVariations=opts.doSystematics,
analysisType="HToTB",
verbose=opts.verbose,
systVarsList=opts.systVarsList)
# Add variations (e.g. for optimisation)
# builder.addVariation("BJetSelection.triggerMatchingApply", [True, False]) # At least 1 trg b-jet dR-matched to offline b-jets
# builder.addVariation("FakeBMeasurement.prelimTopFitChiSqrCutValue", [100, 20])
# builder.addVariation("FakeBMeasurement.prelimTopFitChiSqrCutDirection", ["<=", "==", ">="])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutValue", [0, 1])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutDirection", ["=="])
# builder.addVariation("FakeBMeasurement.numberOfBJetsCutDirection", ["<=", "==", ">="])
# builder.addVariation("FakeBMeasurement.numberOfInvertedBJetsCutValue", [0, 1])
# builder.addVariation("FakeBMeasurement.numberOfInvertedBJetsCutDirection", [">="])
# builder.addVariation("FakeBMeasurement.invertedBJetDiscr", "")
# builder.addVariation("FakeBMeasurement.invertedBJetDiscrWorkingPoint", "Loose")
# builder.addVariation("FakeBMeasurement.invertedBJetsSortType", ["Random", "DescendingBDiscriminator"])
# builder.addVariation("FakeBMeasurement.invertedBJetsDiscrMaxCutValue", [0.82, 0.80, 0.75, 0.70])
# builder.addVariation("TopSelection.ChiSqrCutValue", [100])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT450_SixJet40"], ["HLT_PFHT400_SixJet30"]])
# builder.addVariation("TopologySelection.FoxWolframMomentCutValue", [0.5, 0.7])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056"], ["HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# builder.addVariation("Trigger.triggerOR", [["HLT_PFHT400_SixJet30_DoubleBTagCSV_p056", "HLT_PFHT450_SixJet40_BTagCSV_p056"]])
# Build the builder
builder.build(process, allSelections)
# ================================================================================================
# Example of adding an analyzer whose configuration depends on dataVersion
# ================================================================================================
# def createAnalyzer(dataVersion):
# a = Analyzer("ExampleAnalysis")
# if dataVersion.isMC():
# a.tauPtCut = 10
# else:
# a.tauPtCut = 20
# return a
# process.addAnalyzer("test2", createAnalyzer)
# ================================================================================================
# Pick events
# ================================================================================================
# process.addOptions(EventSaver = PSet(enabled = True,pickEvents = True))
# ================================================================================================
# Run the analysis
# ================================================================================================
# Run the analysis with PROOF? You can give proofWorkers=<N> as a parameter
if opts.jCores:
Print("Running process with PROOF (proofWorkes=%s)" %
(str(opts.jCores)))
process.run(proof=True, proofWorkers=opts.jCores)
else:
Print("Running process (no PROOF)")
process.run()
# Print total time elapsed
tFinish = time.time()
dt = int(tFinish) - int(tStart)
days = divmod(dt, 86400) # days
hours = divmod(days[1], 3600) # hours
mins = divmod(hours[1], 60) # minutes
secs = mins[1] # seconds
Print(
"Total elapsed time is %s days, %s hours, %s mins, %s secs" %
(days[0], hours[0], mins[0], secs), True)
return