def __init__(self,**kwargs):
inputFileNames = kwargs.pop('inputFileNames',[])
inputTreeDirectory = kwargs.pop('inputTreeDirectory','miniTree')
inputTreeName = kwargs.pop('inputTreeName','MiniTree')
inputLumiName = kwargs.pop('inputTreeName','LumiTree')
outputFileName = kwargs.pop('outputFileName','analysisTree.root')
outputTreeName = kwargs.pop('outputTreeName','AnalysisTree')
self.shift = kwargs.pop('shift','')
self.outputTreeName = outputTreeName
if hasProgress:
self.pbar = kwargs.pop('progressbar',ProgressBar(widgets=['{0}: '.format(outputTreeName),' ',SimpleProgress(),' events ',Percentage(),' ',Bar(),' ',ETA()]))
# preselection
if not hasattr(self,'preselection'): self.preselection = '1'
# input files
self.fileNames = []
if os.path.isfile('PSet.py'): # grab input files from crab pset
import PSet
self.fileNames = list(PSet.process.source.fileNames)
elif isinstance(inputFileNames, basestring): # inputFiles is a file name
if os.path.isfile(inputFileNames): # single file
if inputFileNames[-4:] == 'root': # file is a root file
self.fileNames += [inputFileNames]
else: # file is list of files
with open(inputFileNames,'r') as f:
for line in f:
self.fileNames += [line.strip()]
else:
self.fileNames = inputFileNames # already a python list or a cms.untracked.vstring()
if not isinstance(outputFileName, basestring): # its a cms.string(), get value
outputFileName = outputFileName.value()
# test for hdfs
#self.hasHDFS = os.path.exists('/hdfs/store/user')
self.hasHDFS = False
# input tchain
self.treename = '{0}/{1}'.format(inputTreeDirectory,inputTreeName)
luminame = '{0}/{1}'.format(inputTreeDirectory,inputLumiName)
#tchainLumi = ROOT.TChain(luminame)
self.totalEntries = 0
self.numLumis = 0
self.numEvents = 0
self.summedWeights = 0
logging.info('Getting Lumi information')
#self.skims = {}
for f,fName in enumerate(self.fileNames):
if fName.startswith('/store'): fName = '{0}/{1}'.format('/hdfs' if self.hasHDFS else 'root://cmsxrootd.hep.wisc.edu/',fName)
tfile = ROOT.TFile.Open(fName)
tree = tfile.Get(self.treename)
#skimName = 'skim{0}'.format(f)
#tree.Draw('>>{0}'.format(skimName),self.preselection,'entrylist')
#skimlist = ROOT.gDirectory.Get(skimName)
#listEvents = skimlist.GetN()
#self.skims[f] = skimlist
#self.totalEntries += listEvents
self.totalEntries += tree.GetEntries()
if not hasattr(self,'version'):
tree.GetEntry(1)
if hasattr(tree,'provenance'):
ver = tree.provenance[0].split('_')
self.version = ''.join([ver[1],ver[2],'X'])
else:
self.version = getCMSSWVersion()
lumitree = tfile.Get(luminame)
for entry in lumitree:
self.numLumis += 1
self.numEvents += lumitree.nevents
self.summedWeights += lumitree.summedWeights
tfile.Close('R')
#tchainLumi.Add(fName)
# get the lumi info
#self.numLumis = tchainLumi.GetEntries()
#self.numEvents = 0
#self.summedWeights = 0
#for entry in xrange(self.numLumis):
# tchainLumi.GetEntry(entry)
# self.numEvents += tchainLumi.nevents
# self.summedWeights += tchainLumi.summedWeights
logging.info('Analysis is running with version {0}'.format(self.version))
logging.info("Will process {0} lumi sections with {1} events ({2}).".format(self.numLumis,self.numEvents,self.summedWeights))
self.flush()
if not len(self.fileNames): raise Exception
# other input files
self.pileupWeights = PileupWeights(self.version)
self.fakeRates = FakeRates(self.version)
self.leptonScales = LeptonScales(self.version)
self.triggerScales = TriggerScales(self.version)
self.triggerPrescales = TriggerPrescales(self.version)
self.zptGenWeight = ZptGenWeight(self.version)
self.zzGenWeight = ZZGenWeight(self.version)
# tfile
self.outfile = ROOT.TFile(outputFileName,"recreate")
# cut tree
self.cutTree = CutTree()
# analysis tree
self.tree = AnalysisTree(outputTreeName)
self.eventsStored = 0
# some things we always need:
dysamples = [
'DY1JetsToLL_M-50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
'DY2JetsToLL_M-50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
'DY3JetsToLL_M-50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
'DY4JetsToLL_M-50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
'DYJetsToLL_M-10to50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
'DYJetsToLL_M-50_TuneCUETP8M1_13TeV-madgraphMLM-pythia8',
]
qqzzsamples = [
'ZZTo4L_13TeV_powheg_pythia8',
'ZZTo4L_13TeV-amcatnloFXFX-pythia8',
]
# pileup
self.tree.add(lambda cands: self.pileupWeights.weight(self.event)[0], 'pileupWeight', 'F')
self.tree.add(lambda cands: self.pileupWeights.weight(self.event)[1], 'pileupWeightUp', 'F')
self.tree.add(lambda cands: self.pileupWeights.weight(self.event)[2], 'pileupWeightDown', 'F')
self.tree.add(lambda cands: self.event.vertices_count(), 'numVertices', 'I')
self.tree.add(lambda cands: self.event.rho(), 'rho', 'F')
# gen
self.tree.add(lambda cands: self.event.nTrueVertices(), 'numTrueVertices', 'I')
self.tree.add(lambda cands: self.event.NUP(), 'NUP', 'I')
self.tree.add(lambda cands: self.event.isData(), 'isData', 'I')
self.tree.add(lambda cands: self.event.genWeight(), 'genWeight', 'F')
if any([x in fName for x in dysamples]):
self.tree.add(lambda cands: self.zptGenWeight.weight(self.gen), 'zPtWeight', 'F')
if any([x in fName for x in qqzzsamples]):
self.tree.add(lambda cands: self.zzGenWeight.weight(self.gen), 'qqZZkfactor', 'F')
self.tree.add(lambda cands: self.event.numGenJets(), 'numGenJets', 'I')
self.tree.add(lambda cands: self.event.genHT(), 'genHT', 'I')
# scale shifts
weightMap = {
0: {'muR':1.0, 'muF':1.0},
1: {'muR':1.0, 'muF':2.0},
2: {'muR':1.0, 'muF':0.5},
3: {'muR':2.0, 'muF':1.0},
4: {'muR':2.0, 'muF':2.0},
5: {'muR':2.0, 'muF':0.5},
6: {'muR':0.5, 'muF':1.0},
7: {'muR':0.5, 'muF':2.0},
8: {'muR':0.5, 'muF':0.5},
}
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[0] if len(self.event.genWeights())>0 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[0]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[1] if len(self.event.genWeights())>1 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[1]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[2] if len(self.event.genWeights())>2 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[2]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[3] if len(self.event.genWeights())>3 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[3]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[4] if len(self.event.genWeights())>4 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[4]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[5] if len(self.event.genWeights())>5 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[5]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[6] if len(self.event.genWeights())>6 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[6]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[7] if len(self.event.genWeights())>7 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[7]), 'F')
self.tree.add(lambda cands: 0. if self.event.isData() else self.event.genWeights()[8] if len(self.event.genWeights())>8 else 0., 'genWeight_muR{muR:3.1f}_muF{muF:3.1f}'.format(**weightMap[8]), 'F')
def __init__(self, **kwargs):
inputFileNames = kwargs.pop('inputFileNames', [])
inputTreeDirectory = kwargs.pop('inputTreeDirectory', '')
inputTreeName = kwargs.pop('inputTreeName', 'AnalysisTree')
outputFileName = kwargs.pop('outputFileName', 'analysisTree.root')
outputTreeName = kwargs.pop('outputTreeName', 'AnalysisTree')
self.outputTreeName = outputTreeName
if hasProgress:
self.pbar = kwargs.pop(
'progressbar',
ProgressBar(widgets=[
'{0}: '.format(outputTreeName), ' ',
SimpleProgress(), ' events ',
Percentage(), ' ',
Bar(), ' ',
ETA()
]))
# input files
self.fileNames = []
if os.path.isfile('PSet.py'): # grab input files from crab pset
import PSet
self.fileNames = list(PSet.process.source.fileNames)
elif isinstance(inputFileNames,
basestring): # inputFiles is a file name
if os.path.isfile(inputFileNames): # single file
if inputFileNames[-4:] == 'root': # file is a root file
self.fileNames += [inputFileNames]
else: # file is list of files
with open(inputFileNames, 'r') as f:
for line in f:
self.fileNames += [line.strip()]
else:
self.fileNames = inputFileNames # already a python list or a cms.untracked.vstring()
if not isinstance(outputFileName,
basestring): # its a cms.string(), get value
outputFileName = outputFileName.value()
# test for hdfs
#self.hasHDFS = os.path.exists('/hdfs/store/user')
self.hasHDFS = False
# input tchain
self.treename = '{0}/{1}'.format(
inputTreeDirectory,
inputTreeName) if inputTreeDirectory else inputTreeName
self.totalEntries = 0
self.numLumis = 0
self.numEvents = 0
self.summedWeights = 0
logging.info('Getting information')
if len(self.fileNames) == 0: logging.warning('No files to process')
if len(self.fileNames) > 1:
logging.warning(
'More than one file requested, only processing the first file')
for f, fName in enumerate(self.fileNames):
if fName.startswith('/store'):
fName = '{0}/{1}'.format(
'/hdfs' if self.hasHDFS else
'root://cmsxrootd.hep.wisc.edu/', fName)
tfile = ROOT.TFile.Open(fName)
tree = tfile.Get(self.treename)
self.totalEntries += tree.GetEntries()
if not hasattr(self, 'version'):
tree.GetEntry(1)
if hasattr(tree, 'provenance'):
ver = tree.provenance[0].split('_')
self.version = ''.join([ver[1], ver[2], 'X'])
else:
self.version = getCMSSWVersion()
tfile.Close('R')
logging.info('Analysis is running with version {0}'.format(
self.version))
self.flush()
if not len(self.fileNames): raise Exception
# other input files
self.pileupWeights = PileupWeights(self.version)
self.fakeRates = FakeRates(self.version)
self.leptonScales = LeptonScales(self.version)
self.triggerScales = TriggerScales(self.version)
self.triggerPrescales = TriggerPrescales(self.version)
self.zptGenWeight = ZptGenWeight(self.version)
self.zzGenWeight = ZZGenWeight(self.version)
# tfile
fName = self.fileNames[0]
if fName.startswith('/store'):
fName = '{0}/{1}'.format(
'/hdfs' if self.hasHDFS else 'root://cmsxrootd.hep.wisc.edu/',
fName)
self.tfile = ROOT.TFile.Open(fName, 'READ')
self.oldtree = self.tfile.Get(self.treename)
self.outfile = ROOT.TFile(outputFileName, "recreate")
self.tree = self.oldtree.CloneTree(0)
summedWeights = self.tfile.Get('summedWeights')
self.summedWeights = summedWeights.GetBinContent(1)
