import ROOT
import sys, os, json
from TTH.MEAnalysis.samples_base import getSitePrefix
import math
DATASETPATH = os.environ["DATASETPATH"]
FILE_NAMES = os.environ["FILE_NAMES"].split()
ch = ROOT.TChain("tree")
for fi in FILE_NAMES:
ch.AddFile(getSitePrefix(fi))
for ev in ch:
if not (ev.is_sl and ev.numJets==6 and ev.nBCSVM==4):
continue
jets_p4 = []
jets_btag = []
jets_csv = []
jets_cmva = []
jets_matchFlag = []
jets_hadronFlavour = []
for ijet in range(ev.numJets):
p4 = [
ev.jets_pt[ijet],
ev.jets_eta[ijet],
ev.jets_phi[ijet],
ev.jets_mass[ijet]
]
jets_p4 += [p4]
jets_btag += [ev.jets_btagFlag[ijet]]
v["var_types"],
v["tree"],
v["float_branches"],
datatype = 'float')
########################################
# Event loop
########################################
intree = ROOT.TChain("tree")
if "FILE_NAMES" in os.environ.keys():
for fi in os.environ["FILE_NAMES"].split(" "):
print "adding", fi
intree.AddFile(getSitePrefix(fi))
else:
for fi in sys.argv[2:]:
print "adding", fi
intree.AddFile(getSitePrefix(fi))
#intree.AddFile(fi)
n_entries = intree.GetEntries()
print "Will process {0} events".format(n_entries)
for i_event in range(n_entries):
# Progress
if not i_event % 1000:
#find integral of csv distribution in this pt/eta bin
#int_ij = 0.
#for k in range(0, nbinsZ + 2):
# int_ij += h3.GetBinContent(i,j,k)
int_ij = float(h3.ProjectionZ("asd", i, i, j, j).Integral())
#normalize csv histogram
for k in range(0, nbinsZ + 2):
unnorm = float(h3.GetBinContent(i, j, k))
if int_ij > 0.0:
unnorm = unnorm / int_ij
h3.SetBinContent(i, j, k, unnorm)
#print i, j, h3.ProjectionZ("", i, i, j, j).Integral()
if __name__ == "__main__":
tt = ROOT.TChain("vhbb/tree")
for inf in INFILES:
tt.AddFile(getSitePrefix(inf))
of = ROOT.TFile(OUTFILE, "RECREATE")
of.cd()
makeControlPlots("btagCSV", "Jet_btagCSV", 0.0, 1.0)
makeControlPlots("btagCMVA", "Jet_btagCMVA", -1.0, 1.0)
makeControlPlots("btagCMVA_log",
"log((1.0 + Jet_btagCMVA)/(1.0 - Jet_btagCMVA))", -15.0,
15.0)
of.Write()
of.Close()
elif event["is_sl"] and pass_HLT_sl_el(event):
return TRIGGERPATH_MAP["e"]
elif event["is_dl"] and pass_HLT_dl_mumu(event):
return TRIGGERPATH_MAP["mm"]
elif event["is_dl"] and pass_HLT_dl_elmu(event):
return TRIGGERPATH_MAP["em"]
elif event["is_dl"] and pass_HLT_dl_elel(event):
return TRIGGERPATH_MAP["ee"]
return 0
if __name__ == "__main__":
if os.environ.has_key("FILE_NAMES"):
file_names = map(getSitePrefix, os.environ["FILE_NAMES"].split())
prefix, sample = get_prefix_sample(os.environ["DATASETPATH"])
else:
file_names = [getSitePrefix("/store/user/jpata/tth/pilot_Jul30_v1/ttHTobb_M125_13TeV_powheg_pythia8/pilot_Jul30_v1/160730_115048/0000/tree_{0}.root".format(i)) for i in range(1, 10)]
prefix = ""
sample = "ttHTobb_M125_13TeV_powheg_pythia8"
process = samples_nick[sample]
schema = get_schema(sample)
#configure systematic scenarios according to MC/Data
if schema == "mc":
systematics_event = ["nominal"] + SYSTEMATICS_EVENT
systematics_weight = [k[0] for k in systematic_weights]
elif schema == "data":
systematics_event = ["nominal"]
systematics_weight = []
dirs = {}
def Make_config(do_subjets=False):
config = {}
########################################
# Information concerning this config file
########################################
ts = time.time()
config['date'] = datetime.datetime.fromtimestamp(ts).strftime(
'%Y-%m-%d %H:%M:%S')
config['info'] = '*** Information on this config.dat ***\n'\
'This config file contains the configuration data used for outputtree.py,'\
' which translates a VHBB Ntuple to a format readable by readtree.py. \n\n'\
'This config.dat was created on: {0}'.format(config['date'])
########################################
# I/O information
########################################
config['input_tree_name'] = 'vhbb/tree'
#the path to the root files
config["root_file_base"] = samples_base.getSitePrefix()
#config["root_file_base"] = 'file:///hdfs/cms/'
#config["root_file_base"] = "dcap:///pnfs/psi.ch/cms/trivcat/"
# The config file will be copied to 'runs/{config['run_name']}'
config['run_name'] = samples.version
if do_subjets:
config['run_name'] += "_subjet"
else:
config['run_name'] += "_resolved"
config['output_root_file_name'] = 'out.root'
########################################
# Program parameters
########################################
# Use only a part of the input root file
config['Use_limited_entries'] = False
# Specify the number of entries if only a limited number of entries is used
# This number is not used if Use_limited_entries is set to False
config['n_entries_limited'] = 10000
# Specify whether the program should attempt to find MC branches for the jets
config['Get_MC_for_jets'] = False
# Specify whether the program should link with quarks.
# If this is set to True, the program will not save any quark data, and the
# the user should calculate TFs with the MC values of the jets. Make sure
# config['Get_MC_for_jets'] is set to True if this is set to True.
config['Dont_Link_Just_Jets'] = False
# Specify whether the program should look for a closely matched second quark.
# If two quarks could both be matched to a jet, it usually better to remove
# these quarks and this jet altogether, in order to prevent faulty matching.
config['Remove_double_match'] = True
########################################
# Branch info
########################################
# Specify the names of the particles of with pt, eta, phi and m should be
# extracted.
# - pt, eta, phi and mass are extracted by default. A branch E is created by
# default - it is calculated with the use of pt, eta, phi and mass.
# - Since the notation '_pt' is common but not *standard*, it is necessary to
# to add underscores where necessary manually.
config['quarktypes'] = ['GenBQuarkFromTop_', 'GenBQuarkFromH_', 'GenWZQuark_' ]
#enable this for resolved jets
if not do_subjets:
config['jettypes'] = [ 'Jet_' ]
else:
config['jettypes'] = [ 'httCandidates_sjW1', 'httCandidates_sjW2',
'httCandidates_sjNonW'
]
# Specify which branches *other* than pt, eta, phi, mass and E should be
# extracted.
# - This should be FULL branch names, e.g. httCandidates_fW
# - If the extra variable is particle-specific, write '{particle}' in front
# it. For example: '(a quark)pdgId' can be written as '{particle}pdgId'
config['quark_extra_vars'] = [
'{particle}pdgId',
#'{particle}charge',
#'{particle}status',
]
if not do_subjets:
config['jet_extra_vars'] = [
'{particle}hadronFlavour',
'{particle}btagCSV',
'{particle}btagBDT',
]
else:
config['jet_extra_vars'] = [
'httCandidates_pt',
'httCandidates_eta',
'httCandidates_phi',
'httCandidates_mass',
'httCandidates_fRec',
]
########################################
# Cutoff criteria
########################################
# format of 1 cutoff criterium: ( varname, operator sign, cutoff value )
# Note: Only defined variable names can be used here!
config['jet_cutoff_list'] = [
( '{particle}pt' , '>' , 30.0 ),
#( 'httCandidates_pt' , '>' , 200.0 ),
#( 'httCandidates_mass' , '>' , 120.0 ),
#( 'httCandidates_mass' , '<' , 220.0 ),
#( 'httCandidates_fW' , '<' , 0.175 ),
]
config['quark_cutoff_list'] = [
( '{particle}pt' , '>' , 30.0 ) ]
#matching dR between gen and reco
config['max_link_delR'] = 0.3
# Only used if config['Remove_double_match'] is set to True
# if another match closer than dR, remove entire jet
config['max_sec_delR'] = 0.5
########################################
# Write configuration to file:
########################################
f = open( 'cfg_outputtree.dat', 'wb' )
json.dump( config , f, indent=2)
f.close()
if not os.path.isdir( 'runs/{0}'.format(config['run_name'] ) ):
os.makedirs('runs/{0}'.format(config['run_name'] ))
shutil.copyfile( 'cfg_outputtree.dat',
'runs/{0}/cfg_outputtree.dat'.format( config['run_name'] ) )
shutil.copyfile( 'cfg_outputtree.py',
'runs/{0}/cfg_outputtree.py'.format( config['run_name'] ) )
print "cfg_outputtree.dat created"
return TRIGGERPATH_MAP["mm"]
elif event["is_dl"] and pass_HLT_dl_elmu(event):
return TRIGGERPATH_MAP["em"]
elif event["is_dl"] and pass_HLT_dl_elel(event):
return TRIGGERPATH_MAP["ee"]
return 0
if __name__ == "__main__":
if os.environ.has_key("FILE_NAMES"):
file_names = map(getSitePrefix, os.environ["FILE_NAMES"].split())
prefix, sample = get_prefix_sample(os.environ["DATASETPATH"])
else:
file_names = [
getSitePrefix(
"/store/user/jpata/tth/pilot_Jul30_v1/ttHTobb_M125_13TeV_powheg_pythia8/pilot_Jul30_v1/160730_115048/0000/tree_{0}.root"
.format(i)) for i in range(1, 10)
]
prefix = ""
sample = "ttHTobb_M125_13TeV_powheg_pythia8"
process = samples_nick[sample]
schema = get_schema(sample)
#configure systematic scenarios according to MC/Data
if schema == "mc":
systematics_event = ["nominal"] + SYSTEMATICS_EVENT
systematics_weight = [k[0] for k in systematic_weights]
elif schema == "data":
systematics_event = ["nominal"]
systematics_weight = []
import ROOT
import sys
from TTH.MEAnalysis.samples_base import getSitePrefix
ofname = sys.argv[1]
tt = ROOT.TChain("tree")
for fi in sys.argv[2:]:
print "adding", fi
fn = getSitePrefix(fi)
tf = ROOT.TFile.Open(fn)
if not tf or tf.IsZombie():
raise Exception("Could not open file: {0}".format(fn))
tf.Close()
tt.AddFile(fn)
tt.SetBranchStatus("*", False)
tt.SetBranchStatus("mem_tt*", True)
tt.SetBranchStatus("nMatch*", True)
tt.SetBranchStatus("is_*", True)
tt.SetBranchStatus("numJets*", True)
tt.SetBranchStatus("nB*", True)
tt.SetBranchStatus("n_*", True)
#tt.SetBranchStatus("topCand*", True)
tt.SetBranchStatus("btag_LR_4b_2b*", True)
tt.SetBranchStatus("ttCls", True)
tt.SetBranchStatus("run", True)
tt.SetBranchStatus("lumi", True)
tt.SetBranchStatus("evt", True)
tt.SetBranchStatus("cat", True)
#tt.SetBranchStatus("ht", True)
#tt.SetBranchStatus("isotropy", True)
import ROOT
import sys
from TTH.MEAnalysis.samples_base import getSitePrefix
ofname = sys.argv[1]
tt = ROOT.TChain("vhbb/tree")
for fi in sys.argv[2:]:
print "adding", fi
tt.AddFile(getSitePrefix(fi))
tt.SetBranchStatus("*", False)
tt.SetBranchStatus("GenBQuarkFromH*", True)
tt.SetBranchStatus("GenBQuarkFromTop*", True)
tt.SetBranchStatus("GenGluon*", True)
tt.SetBranchStatus("GenHiggsBoson*", True)
tt.SetBranchStatus("GenJet*", True)
tt.SetBranchStatus("Jet*", True)
tt.SetBranchStatus("aLeptons*", True)
tt.SetBranchStatus("GenLep*", True)
tt.SetBranchStatus("GenLepFromTop*", True)
tt.SetBranchStatus("GenNuFromTop*", True)
tt.SetBranchStatus("GenTop*", True)
tt.SetBranchStatus("GenWZQuark*", True)
tt.SetBranchStatus("GenStatus2bHad*", True)
tt.SetBranchStatus("ttCls*", True)
of = ROOT.TFile(ofname, "RECREATE")
of.cd()
tt.CopyTree("1")
of.Write()
of.Close()
for i in range(0, nbinsX+2):
for j in range(0, nbinsY+2):
#find integral of csv distribution in this pt/eta bin
#int_ij = 0.
#for k in range(0, nbinsZ + 2):
# int_ij += h3.GetBinContent(i,j,k)
int_ij = float(h3.ProjectionZ("asd", i, i, j, j).Integral())
#normalize csv histogram
for k in range(0, nbinsZ + 2):
unnorm = float(h3.GetBinContent(i,j,k))
if int_ij > 0.0:
unnorm = unnorm / int_ij
h3.SetBinContent(i, j, k, unnorm)
#print i, j, h3.ProjectionZ("", i, i, j, j).Integral()
if __name__ == "__main__":
tt = ROOT.TChain("vhbb/tree")
for inf in INFILES:
tt.AddFile(getSitePrefix(inf))
of = ROOT.TFile(OUTFILE, "RECREATE")
of.cd()
makeControlPlots("btagCSV", "Jet_btagCSV", 0.0, 1.0)
makeControlPlots("btagCMVA", "Jet_btagCMVA", -1.0, 1.0)
makeControlPlots("btagCMVA_log", "log((1.0 + Jet_btagCMVA)/(1.0 - Jet_btagCMVA))", -15.0, 15.0)
of.Write()
of.Close()
import ROOT, sys
from TTH.MEAnalysis.samples_base import getSitePrefix
tf = ROOT.TFile.Open(getSitePrefix(sys.argv[1]))
if not tf:
raise Exception("Could not open file")
tt = tf.Get(sys.argv[2])
branches = sorted([br.GetName() for br in tt.GetListOfBranches()])
for br in branches:
print br
AH.addScalarBranches(v["vars"],
v["var_types"],
v["tree"],
v["float_branches"],
datatype='float')
########################################
# Event loop
########################################
intree = ROOT.TChain("tree")
if "FILE_NAMES" in os.environ.keys():
for fi in os.environ["FILE_NAMES"].split(" "):
print "adding", fi
intree.AddFile(getSitePrefix(fi))
else:
for fi in sys.argv[2:]:
print "adding", fi
intree.AddFile(getSitePrefix(fi))
#intree.AddFile(fi)
n_entries = intree.GetEntries()
print "Will process {0} events".format(n_entries)
for i_event in range(n_entries):
# Progress
if not i_event % 1000:
print "{0:.1f}%".format(100. * i_event / n_entries)
import ROOT
import sys
from TTH.MEAnalysis.samples_base import getSitePrefix
ofname = sys.argv[1]
tt = ROOT.TChain("tree")
for fi in sys.argv[2:]:
print "adding", fi
fn = getSitePrefix(fi)
tf = ROOT.TFile.Open(fn)
if not tf or tf.IsZombie():
raise Exception("Could not open file: {0}".format(fn))
tf.Close()
tt.AddFile(fn)
tt.SetBranchStatus("*", False)
tt.SetBranchStatus("mem_tt*", True)
tt.SetBranchStatus("nMatch*", True)
tt.SetBranchStatus("is_*", True)
tt.SetBranchStatus("numJets*", True)
tt.SetBranchStatus("nB*", True)
tt.SetBranchStatus("n_*", True)
#tt.SetBranchStatus("topCand*", True)
tt.SetBranchStatus("btag_LR_4b_2b*", True)
tt.SetBranchStatus("ttCls", True)
tt.SetBranchStatus("run", True)
tt.SetBranchStatus("lumi", True)
tt.SetBranchStatus("evt", True)
tt.SetBranchStatus("cat", True)
#tt.SetBranchStatus("ht", True)
#tt.SetBranchStatus("isotropy", True)