def buildTriggerCount(selectionParams,sampleSelectionParams,projectionParams,sampleProjectionParams,histParams,sampleHistParams):
histParams['TriggerCount'] = {
'count' : {'xVariable': '1', 'xBinning': [1,0,2], }, # just a count of events passing selection
}
triggers = [
'pass_Mu17_TrkIsoVVL_Mu8_TrkIsoVVL_DZ',
'pass_Mu17_TrkIsoVVL_TkMu8_TrkIsoVVL_DZ',
'pass_Ele23_Ele12_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_IsoMu24',
'pass_IsoTkMu24',
'pass_Ele27_WPTight_Gsf',
'pass_Mu8_TrkIsoVVL_Ele23_CaloIdL_TrackIdL_IsoVL',
'pass_Mu23_TrkIsoVVL_Ele8_CaloIdL_TrackIdL_IsoVL',
'pass_Mu8_TrkIsoVVL_Ele23_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_Mu23_TrkIsoVVL_Ele8_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_DoubleMediumCombinedIsoPFTau35_Trk1_eta2p1_Reg',
]
for trigger in triggers:
histParams['TriggerCount'][trigger] = {'xVariable': trigger, 'xBinning': [2,0,2], }
selectionParams['TriggerCount'] = {
'all' : {'args': ['1'], 'kwargs': {'mcscalefactor': '1', }},
}
intLumi = getLumi()
for r in ['B','C','D','E','F','G','H']:
run = 'Run2016{0}'.format(r)
runLumi = getLumi(run=run)
mcscale = '{0}/{1}'.format(runLumi,intLumi)
runRange = getRunRange(run=run)
datacut = 'run>={0} && run<={1}'.format(*runRange)
selectionParams['TriggerCount'][run] = {'args': [datacut], 'kwargs': {'mcscalefactor': mcscale}}
def __init__(self,version):
self.version = version
self.prescales = {}
self.lumi = {
'76X': getLumi(version='76X'),
'80X': getLumi(version='80X'),
}
self.prescales['76X'] = {
# HLT_Ele12_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg2': 4.174,
# HLT_Ele17_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg1': 45.941,
# HLT_Mu8_TrkIsoVVL_v*
'Mu17_Mu8Leg2' : 1.330,
# HLT_Mu17_TrkIsoVVL_v*
'Mu17_Mu8Leg1' : 197.362,
}
self.prescales['80X'] = {
# HLT_Ele12_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg2': 6.162,
'Ele23_Ele12Leg2': 6.162,
# HLT_Ele17_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg1': 30.397,
# HLT_Ele23_CaloIdL_TrackIdL_IsoVL_v*
'Ele23_Ele12Leg1': 16.430,
# HLT_Mu8_TrkIsoVVL_v*
'Mu17_Mu8Leg2' : 7.832,
# HLT_Mu17_TrkIsoVVL_v*
'Mu17_Mu8Leg1' : 217.553,
}
def __init__(self, version):
self.version = version
self.prescales = {}
self.lumi = {
'76X': getLumi(version='76X'),
'80X': getLumi(version='80X'),
}
self.prescales['76X'] = {
# HLT_Ele12_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg2': 4.174,
# HLT_Ele17_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg1': 45.941,
# HLT_Mu8_TrkIsoVVL_v*
'Mu17_Mu8Leg2': 1.330,
# HLT_Mu17_TrkIsoVVL_v*
'Mu17_Mu8Leg1': 197.362,
}
self.prescales['80X'] = {
# HLT_Ele12_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg2': 6.162,
'Ele23_Ele12Leg2': 6.162,
# HLT_Ele17_CaloIdL_TrackIdL_IsoVL_v*
'Ele17_Ele12Leg1': 30.397,
# HLT_Ele23_CaloIdL_TrackIdL_IsoVL_v*
'Ele23_Ele12Leg1': 16.430,
# HLT_Mu8_TrkIsoVVL_v*
'Mu17_Mu8Leg2': 7.832,
# HLT_Mu17_TrkIsoVVL_v*
'Mu17_Mu8Leg1': 217.553,
}
def main(argv=None):
if argv is None:
argv = sys.argv[1:]
args = parse_command_line(argv)
if args.verbose and args.analysis:
table = PrettyTable(['Sample','xsec [pb]','entries','ratio neg.','lumi. [/pb]','eff. entries'])
else:
table = PrettyTable(['Sample','xsec [pb]'])
table.align = 'r'
table.align['Sample'] = 'l'
ntupleDir = getAnalysisNtupleDirectory(args.analysis,True) if args.verbose and args.analysis else getNtupleDirectory(version=args.version)
#Odd mix of local pathnames and xrootd access
# for sample in sorted(hdfs_ls_directory(ntupleDir)):
for sample in (glob.glob('/'.join([ntupleDir,'*']))):
name = os.path.basename(sample)
logging.info('Processing {0}'.format(name))
data = isData(name)
xsec = getXsec(name)
if args.verbose and args.analysis:
print sample
fnames = get_hdfs_root_files(sample)
# get total events, total weights
tree = ROOT.TChain(getTreeName(args.analysis))
summedWeights = 0.
for f in fnames:
tfile = ROOT.TFile.Open('/hdfs'+f)
summedWeights += tfile.Get("summedWeights").GetBinContent(1)
tfile.Close()
tree.Add('/hdfs'+f)
numEntries = tree.GetEntries(args.selection)
weightedEntries = 0.
negevents = 0.
seltree = tree.CopyTree(args.selection)
for row in seltree:
if data:
weightedEntries += 1.
else:
weightedEntries += row.genWeight
if row.genWeight<0.: negevents += 1
if data:
sampleLumi = getLumi()
else:
sampleLumi = float(summedWeights)/xsec if xsec else 0.
negratio = float(negevents)/numEntries if numEntries else 0.
effevents = weightedEntries*getLumi()/sampleLumi if sampleLumi else 0.
table.add_row([name,'{0:.6f}'.format(float(xsec)),numEntries,'{0:.3}'.format(float(negratio)),'{0:.6f}'.format(float(sampleLumi)),'{0:.3f}'.format(float(effevents))])
else:
table.add_row([name,xsec])
print table.get_string()
def main(argv=None):
if argv is None:
argv = sys.argv[1:]
args = parse_command_line(argv)
if args.verbose and args.analysis:
table = PrettyTable(['Sample','xsec [pb]','entries','ratio neg.','lumi. [/pb]','eff. entries'])
else:
table = PrettyTable(['Sample','xsec [pb]'])
table.align = 'r'
table.align['Sample'] = 'l'
ntupleDir = getAnalysisNtupleDirectory(args.analysis) if args.verbose and args.analysis else getNtupleDirectory(version=args.version)
for sample in sorted(glob.glob(os.path.join(ntupleDir,'*'))):
name = os.path.basename(sample)
logging.info('Processing {0}'.format(name))
data = isData(name)
xsec = getXsec(name)
if args.verbose and args.analysis:
fnames = get_hdfs_root_files(sample)
# get total events, total weights
tree = ROOT.TChain(getTreeName(args.analysis))
summedWeights = 0.
for f in fnames:
tfile = ROOT.TFile.Open('/hdfs'+f)
summedWeights += tfile.Get("summedWeights").GetBinContent(1)
tfile.Close()
tree.Add('/hdfs'+f)
numEntries = tree.GetEntries(args.selection)
weightedEntries = 0.
negevents = 0.
seltree = tree.CopyTree(args.selection)
for row in seltree:
if data:
weightedEntries += 1.
else:
weightedEntries += row.genWeight
if row.genWeight<0.: negevents += 1
if data:
sampleLumi = getLumi()
else:
sampleLumi = float(summedWeights)/xsec if xsec else 0.
negratio = float(negevents)/numEntries if numEntries else 0.
effevents = weightedEntries*getLumi()/sampleLumi if sampleLumi else 0.
table.add_row([name,'{0:.6f}'.format(float(xsec)),numEntries,'{0:.3}'.format(float(negratio)),'{0:.6f}'.format(float(sampleLumi)),'{0:.3f}'.format(float(effevents))])
else:
table.add_row([name,xsec])
print table.get_string()
def buildTriggerCount(selectionParams, sampleSelectionParams, projectionParams,
sampleProjectionParams, histParams, sampleHistParams):
histParams['TriggerCount'] = {
'count': {
'xVariable': '1',
'xBinning': [1, 0, 2],
}, # just a count of events passing selection
}
triggers = [
'pass_Mu17_TrkIsoVVL_Mu8_TrkIsoVVL_DZ',
'pass_Mu17_TrkIsoVVL_TkMu8_TrkIsoVVL_DZ',
'pass_Ele23_Ele12_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_IsoMu24',
'pass_IsoTkMu24',
'pass_Ele27_WPTight_Gsf',
'pass_Mu8_TrkIsoVVL_Ele23_CaloIdL_TrackIdL_IsoVL',
'pass_Mu23_TrkIsoVVL_Ele8_CaloIdL_TrackIdL_IsoVL',
'pass_Mu8_TrkIsoVVL_Ele23_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_Mu23_TrkIsoVVL_Ele8_CaloIdL_TrackIdL_IsoVL_DZ',
'pass_DoubleMediumCombinedIsoPFTau35_Trk1_eta2p1_Reg',
]
for trigger in triggers:
histParams['TriggerCount'][trigger] = {
'xVariable': trigger,
'xBinning': [2, 0, 2],
}
selectionParams['TriggerCount'] = {
'all': {
'args': ['1'],
'kwargs': {
'mcscalefactor': '1',
}
},
}
intLumi = getLumi()
for r in ['B', 'C', 'D', 'E', 'F', 'G', 'H']:
run = 'Run2016{0}'.format(r)
runLumi = getLumi(run=run)
mcscale = '{0}/{1}'.format(runLumi, intLumi)
runRange = getRunRange(run=run)
datacut = 'run>={0} && run<={1}'.format(*runRange)
selectionParams['TriggerCount'][run] = {
'args': [datacut],
'kwargs': {
'mcscalefactor': mcscale
}
}
def flatten(analysis,sample,**kwargs):
histParams = kwargs.pop('histParams',{})
histSelections = kwargs.pop('histSelections',{})
inputFileList = kwargs.pop('inputFileList','')
outputFile = kwargs.pop('outputFile','')
shift = kwargs.pop('shift','')
countOnly = kwargs.pop('countOnly',False)
njobs = kwargs.pop('njobs',1)
job = kwargs.pop('job',0)
multi = kwargs.pop('multi',False)
useProof = kwargs.pop('useProof',False)
intLumi = kwargs.pop('intLumi',float(getLumi()))
if hasProgress and multi:
pbar = kwargs.pop('progressbar',ProgressBar(widgets=['{0}: '.format(sample),' ',SimpleProgress(),' histograms ',Percentage(),' ',Bar(),' ',ETA()]))
else:
pbar = None
if outputFile:
flat = outputFile
proj = outputFile.replace('.root','_projection.root')
flattener = FlattenTree(analysis,sample,inputFileList=inputFileList,flat=flat,proj=proj,shift=shift,countOnly=countOnly,useProof=useProof,intLumi=intLumi)
else:
flattener = FlattenTree(analysis,sample,inputFileList=inputFileList,shift=shift,countOnly=countOnly,useProof=useProof,intLumi=intLumi)
for histName, params in histParams.iteritems():
flattener.addHistogram(histName,**params)
for selName, sel in histSelections.iteritems():
if sel: flattener.addSelection(selName,**sel['kwargs'])
flattener.flattenAll(progressbar=pbar,njobs=njobs,job=job,multi=multi)
def __initializeNtuple(self):
tchain = ROOT.TChain(self.treeName)
if self.inputFileList: # reading from a passed list of inputfiles
allFiles = []
with open(self.inputFileList, 'r') as f:
for line in f.readlines():
allFiles += [line.strip()]
else: # reading from an input directory (all files in directory will be processed)
allFiles = glob.glob('{0}/*.root'.format(self.ntupleDirectory))
if len(allFiles) == 0:
logging.error('No files found for sample {0}'.format(self.sample))
summedWeights = 0.
for f in allFiles:
tfile = ROOT.TFile.Open(f)
summedWeights += tfile.Get("summedWeights").GetBinContent(1)
tfile.Close()
tchain.Add(f)
if not summedWeights and not isData(self.sample):
logging.warning('No events for sample {0}'.format(self.sample))
self.intLumi = float(getLumi())
self.xsec = getXsec(self.sample)
self.sampleLumi = float(summedWeights) / self.xsec if self.xsec else 0.
self.sampleTree = tchain
self.files = allFiles
self.initialized = True
logging.debug(
'Initialized {0}: summedWeights = {1}; xsec = {2}; sampleLumi = {3}; intLumi = {4}'
.format(self.sample, summedWeights, self.xsec, self.sampleLumi,
self.intLumi))
def __init__(self,analysis,sample,**kwargs):
# default to access via sample/analysis
self.analysis = analysis
self.sample = sample
self.shift = kwargs.pop('shift','')
self.skipHists = kwargs.pop('skipHists',False)
self.isData = isData(self.sample)
self.intLumi = kwargs.get('intLumi',float(getLumi()))
logging.debug('Initializing {0} {1} {2}'.format(self.analysis,self.sample,self.shift))
# backup passing custom parameters
self.ntupleDirectory = kwargs.pop('ntupleDirectory','{0}/{1}'.format(getNtupleDirectory(self.analysis,shift=self.shift),self.sample))
self.inputFileList = kwargs.pop('inputFileList','')
self.outputFile = kwargs.pop('outputFile',getNewFlatHistograms(self.analysis,self.sample,shift=self.shift))
if os.path.dirname(self.outputFile): python_mkdir(os.path.dirname(self.outputFile))
self.treeName = kwargs.pop('treeName',getTreeName(self.analysis))
if hasProgress:
self.pbar = kwargs.pop('progressbar',ProgressBar(widgets=['{0}: '.format(sample),' ',SimpleProgress(),' ',Percentage(),' ',Bar(),' ',ETA()]))
else:
self.pbar = None
# get stuff needed to flatten
self.infile = 0
self.tchain = 0
self.initialized = False
self.hists = {}
self.datasets = {}
def __init__(self, analysis, **kwargs):
'''Initialize the plotter'''
# plot directory
self.analysis = analysis
self.intLumi = kwargs.get('intLumi', float(getLumi()))
self.outputDirectory = kwargs.pop('outputDirectory',
'plots/{0}'.format(self.analysis))
self.outputDirectoryCSV = kwargs.pop('outputDirectoryCSV',
'csv/{0}'.format(self.analysis))
def __init__(self,**kwargs):
self.ntupleDirectory = kwargs.pop('ntupleDirectory','ntuples/Analysis')
self.treeName = kwargs.pop('treeName','AnalysisTree')
# dictionaries to hold the sample information
self.intLumi = getLumi()
self.sampleLumi = 0
self.sampleTree = None
self.histParameters = {}
self.histParameters2D = {}
self.outfile = 0
self.preinitialized = False
self.selections = []
self.sample = ''
self.currVal = 0
def __init__(self, **kwargs):
self.ntupleDirectory = kwargs.pop('ntupleDirectory',
'ntuples/Analysis')
self.treeName = kwargs.pop('treeName', 'AnalysisTree')
# dictionaries to hold the sample information
self.intLumi = getLumi()
self.sampleLumi = 0
self.sampleTree = None
self.histParameters = {}
self.histParameters2D = {}
self.outfile = 0
self.preinitialized = False
self.selections = []
self.sample = ''
self.currVal = 0
def parse_command_line(argv):
parser = argparse.ArgumentParser(description='Flatten Tree')
parser.add_argument('analysis', type=str, help='Analysis to process')
parser.add_argument('shift', type=str, default='', nargs='?', help='Shift to apply to scale factors')
parser.add_argument('countOnly', type=int, default=0, nargs='?', help='Only do counts, no distributions')
parser.add_argument('--samples', nargs='+', type=str, default=['*'], help='Samples to flatten. Supports unix style wildcards.')
parser.add_argument('--hists', nargs='+', type=str, default=['all'], help='Histograms to flatten.')
parser.add_argument('--selections', nargs='+', type=str, default=['all'], help='Selections to flatten.')
parser.add_argument('--channels', nargs='+', type=str, default=['all'], help='Channels to project.')
parser.add_argument('--skipProjection', action='store_true', help='Skip projecting')
parser.add_argument('--intLumi', type=float, default=float(getLumi()), nargs='?', help='Override luminosity from full run')
#parser.add_argument('--useProof', action='store_true', help='Use PROOF')
parser.add_argument('-j',type=int,default=1,help='Number of cores to use')
return parser.parse_args(argv)
def __setStyle(self,pad,position=11,preliminary=True):
'''Set style for plots based on the CMS TDR style guidelines.
https://twiki.cern.ch/twiki/bin/view/CMS/Internal/PubGuidelines#Figures_and_tables
https://ghm.web.cern.ch/ghm/plots/'''
# set period (used in CMS_lumi)
# period : sqrts
# 1 : 7, 2 : 8, 3 : 7+8, 4 : 13, ... 7 : 7+8+13
period_int = 4
# set position
# 11: upper left, 33 upper right
CMS_lumi.wrtieExtraText = preliminary
CMS_lumi.extraText = "Preliminary"
CMS_lumi.lumi_13TeV = "%0.1f fb^{-1}" % (float(getLumi())/1000.)
if getLumi < 1000:
CMS_lumi.lumi_13TeV = "%0.1f pb^{-1}" % (float(getLumi))
CMS_lumi.CMS_lumi(pad,period_int,position)
def _setStyle(self,pad,position=11,preliminary=True,personal=False,period_int=4):
'''Set style for plots based on the CMS TDR style guidelines.
https://twiki.cern.ch/twiki/bin/view/CMS/Internal/PubGuidelines#Figures_and_tables
https://ghm.web.cern.ch/ghm/plots/'''
# set period (used in CMS_lumi)
# period : sqrts
# 1 : 7, 2 : 8, 3 : 7+8, 4 : 13, ... 7 : 7+8+13
# set position
# 11: upper left, 33 upper right
CMS_lumi.cmsText = 'CMS' if not personal else 'Devin N. Taylor'
CMS_lumi.writeExtraText = preliminary if not personal else True
CMS_lumi.extraText = "Preliminary" if not personal else 'Analysis in Progress'
CMS_lumi.lumi_13TeV = "%0.1f fb^{-1}" % (float(getLumi())/1000.)
if getLumi < 1000:
CMS_lumi.lumi_13TeV = "%0.1f pb^{-1}" % (float(getLumi))
CMS_lumi.CMS_lumi(pad,period_int,position)
def __init__(self,analysis,sample,**kwargs):
# default to access via sample/analysis
self.analysis = analysis
self.sample = sample
self.shift = kwargs.pop('shift','')
self.intLumi = kwargs.get('intLumi',float(getLumi()))
logging.debug('Initializing {0} {1} {2}'.format(self.analysis,self.sample,self.shift))
# backup passing custom parameters
self.ntupleDirectory = kwargs.pop('ntupleDirectory','{0}/{1}'.format(getNtupleDirectory(self.analysis,shift=self.shift),self.sample))
self.inputFileList = kwargs.pop('inputFileList','')
self.outputFile = kwargs.pop('outputFile','')
self.json = kwargs.pop('json',getSkimJson(self.analysis,self.sample))
self.pickle = kwargs.pop('pickle',getSkimPickle(self.analysis,self.sample))
self.treeName = kwargs.pop('treeName',getTreeName(self.analysis))
if hasProgress:
self.pbar = kwargs.pop('progressbar',ProgressBar(widgets=['{0}: '.format(sample),' ',SimpleProgress(),' events ',Percentage(),' ',Bar(),' ',ETA()]))
else:
self.pbar = None
# get stuff needed to flatten
self.infile = 0
self.tchain = 0
self.initialized = False
self.counts = {}
def __init__(self, **kwargs):
inputTreeName = kwargs.pop('inputTreeName', 'WZTree')
super(WZTrainer, self).__init__(**kwargs)
sampleDir = 'ntuples/WZ'
sampleMap = {
"dy10":
"DYJetsToLL_M-10to50_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"dy50": "DYJetsToLL_M-50_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ggzz2e2m": "GluGluToContinToZZTo2e2mu_13TeV_MCFM701_pythia8",
"ggzz2e2t": "GluGluToContinToZZTo2e2tau_13TeV_MCFM701_pythia8",
"ggzz2m2t": "GluGluToContinToZZTo2mu2tau_13TeV_MCFM701_pythia8",
"ggzz4e": "GluGluToContinToZZTo4e_13TeV_MCFM701_pythia8",
"ggzz4m": "GluGluToContinToZZTo4mu_13TeV_MCFM701_pythia8",
"ggzz4t": "GluGluToContinToZZTo4tau_13TeV_MCFM701_pythia8",
"tt": "TTJets_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ttw":
"TTWJetsToLNu_TuneCUETP8M1_13TeV-amcatnloFXFX-madspin-pythia8",
"w": "WJetsToLNu_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ww": "WWTo2L2Nu_13TeV-powheg",
"wz3lnu": "WZTo3LNu_TuneCUETP8M1_13TeV-powheg-pythia8",
"wz2l2q": "WZTo2L2Q_13TeV_amcatnloFXFX_madspin_pythia8",
"wzz": "WZZ_TuneCUETP8M1_13TeV-amcatnlo-pythia8",
"zg": "ZGTo2LG_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"zz2l2n": "ZZTo2L2Nu_13TeV_powheg_pythia8",
"zz2l2q": "ZZTo2L2Q_13TeV_amcatnloFXFX_madspin_pythia8",
"zz4l": "ZZTo4L_13TeV_powheg_pythia8",
#"tzq" : "tZq_ll_4f_13TeV-amcatnlo-pythia8_TuneCUETP8M1",
}
# get the trees
intLumis = {}
for s in sampleMap:
summedWeights = 0.
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir,
sampleMap[s])):
tfile = ROOT.TFile.Open(f)
hist = tfile.Get('summedWeights')
summedWeights += hist.GetBinContent(1)
tfile.Close()
intLumis[s] = float(summedWeights) / getXsec(sampleMap[s])
sigTrees = {}
for sig in ['wz3lnu']:
sigTrees[sig] = ROOT.TChain('WZTree')
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir,
sampleMap[sig])):
sigTrees[sig].Add(f)
bgTrees = {}
for bg in [
'dy10', 'dy50', 'ggzz2e2m', 'ggzz2m2t', 'ggzz4e', 'ggzz4m',
'ggzz4t', 'tt', 'ttw', 'wzz', 'zg', 'zz2l2n', 'zz2l2q', 'zz4l'
]:
bgTrees[bg] = ROOT.TChain('WZTree')
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir,
sampleMap[bg])):
bgTrees[bg].Add(f)
intLumi = getLumi()
# add to factory
for sig in sigTrees:
self.factory.AddSignalTree(sigTrees[sig], intLumi / intLumis[sig])
for bg in bgTrees:
self.factory.AddBackgroundTree(bgTrees[bg], intLumi / intLumis[bg])
# per event weight
weight = 'genWeight'
self.factory.SetWeightExpression(weight)
# variables
self.factory.AddVariable('z1_pt', 'F')
self.factory.AddVariable('z2_pt', 'F')
self.factory.AddVariable('w1_pt', 'F')
self.factory.AddVariable('z_mass', 'F')
self.factory.AddVariable('met_pt', 'F')
self.factory.AddVariable('numBjetsTight30', 'I')
# preselection cut
passCut = ROOT.TCut(
'z1_passMedium==1 && z2_passMedium==1 && w1_passTight==1')
self.factory.PrepareTrainingAndTestTree(
passCut, ":".join([
"nTrain_Signal=0", "nTrain_Background=0", "SplitMode=Random",
"NormMode=NumEvents", "!V"
]))
# options:
# H : display help
# V : turn on verbosity
# IgnoreNegWeightsInTraining : ignore events with negative weights for training, keep for testing
# book method
method = self.factory.BookMethod(
ROOT.TMVA.Types.kBDT, "BDT", ":".join([
"NTrees=850",
"MaxDepth=3",
"BoostType=AdaBoost",
"AdaBoostBeta=0.5",
"SeparationType=GiniIndex",
"nCuts=20",
"PruneMethod=NoPruning",
]))
def __init__(self,**kwargs):
inputTreeName = kwargs.pop('inputTreeName','WZTree')
super(WZTrainer,self).__init__(**kwargs)
sampleDir = 'ntuples/WZ'
sampleMap = {
"dy10" : "DYJetsToLL_M-10to50_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"dy50" : "DYJetsToLL_M-50_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ggzz2e2m" : "GluGluToContinToZZTo2e2mu_13TeV_MCFM701_pythia8",
"ggzz2e2t" : "GluGluToContinToZZTo2e2tau_13TeV_MCFM701_pythia8",
"ggzz2m2t" : "GluGluToContinToZZTo2mu2tau_13TeV_MCFM701_pythia8",
"ggzz4e" : "GluGluToContinToZZTo4e_13TeV_MCFM701_pythia8",
"ggzz4m" : "GluGluToContinToZZTo4mu_13TeV_MCFM701_pythia8",
"ggzz4t" : "GluGluToContinToZZTo4tau_13TeV_MCFM701_pythia8",
"tt" : "TTJets_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ttw" : "TTWJetsToLNu_TuneCUETP8M1_13TeV-amcatnloFXFX-madspin-pythia8",
"w" : "WJetsToLNu_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"ww" : "WWTo2L2Nu_13TeV-powheg",
"wz3lnu" : "WZTo3LNu_TuneCUETP8M1_13TeV-powheg-pythia8",
"wz2l2q" : "WZTo2L2Q_13TeV_amcatnloFXFX_madspin_pythia8",
"wzz" : "WZZ_TuneCUETP8M1_13TeV-amcatnlo-pythia8",
"zg" : "ZGTo2LG_TuneCUETP8M1_13TeV-amcatnloFXFX-pythia8",
"zz2l2n" : "ZZTo2L2Nu_13TeV_powheg_pythia8",
"zz2l2q" : "ZZTo2L2Q_13TeV_amcatnloFXFX_madspin_pythia8",
"zz4l" : "ZZTo4L_13TeV_powheg_pythia8",
#"tzq" : "tZq_ll_4f_13TeV-amcatnlo-pythia8_TuneCUETP8M1",
}
# get the trees
intLumis = {}
for s in sampleMap:
summedWeights = 0.
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir, sampleMap[s])):
tfile = ROOT.TFile.Open(f)
hist = tfile.Get('summedWeights')
summedWeights += hist.GetBinContent(1)
tfile.Close()
intLumis[s] = float(summedWeights)/getXsec(sampleMap[s])
sigTrees = {}
for sig in ['wz3lnu']:
sigTrees[sig] = ROOT.TChain('WZTree')
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir,sampleMap[sig])):
sigTrees[sig].Add(f)
bgTrees = {}
for bg in ['dy10','dy50','ggzz2e2m','ggzz2m2t','ggzz4e','ggzz4m','ggzz4t','tt','ttw','wzz','zg','zz2l2n','zz2l2q','zz4l']:
bgTrees[bg] = ROOT.TChain('WZTree')
for f in glob.glob('{0}/{1}/*.root'.format(sampleDir,sampleMap[bg])):
bgTrees[bg].Add(f)
intLumi = getLumi()
# add to factory
for sig in sigTrees:
self.factory.AddSignalTree(sigTrees[sig],intLumi/intLumis[sig])
for bg in bgTrees:
self.factory.AddBackgroundTree(bgTrees[bg],intLumi/intLumis[bg])
# per event weight
weight = 'genWeight'
self.factory.SetWeightExpression(weight)
# variables
self.factory.AddVariable('z1_pt','F')
self.factory.AddVariable('z2_pt','F')
self.factory.AddVariable('w1_pt','F')
self.factory.AddVariable('z_mass','F')
self.factory.AddVariable('met_pt','F')
self.factory.AddVariable('numBjetsTight30','I')
# preselection cut
passCut = ROOT.TCut('z1_passMedium==1 && z2_passMedium==1 && w1_passTight==1')
self.factory.PrepareTrainingAndTestTree(
passCut,
":".join(
[
"nTrain_Signal=0",
"nTrain_Background=0",
"SplitMode=Random",
"NormMode=NumEvents",
"!V"
]
)
)
# options:
# H : display help
# V : turn on verbosity
# IgnoreNegWeightsInTraining : ignore events with negative weights for training, keep for testing
# book method
method = self.factory.BookMethod(
ROOT.TMVA.Types.kBDT,
"BDT",
":".join(
[
"NTrees=850",
"MaxDepth=3",
"BoostType=AdaBoost",
"AdaBoostBeta=0.5",
"SeparationType=GiniIndex",
"nCuts=20",
"PruneMethod=NoPruning",
]
)
)
