def finalizeOutput(job_script_dir, output_dir, flat_ntuples):
# first, check the output is all there
pass_check_output = CheckOutput.checkOutput(job_script_dir, output_dir)
print pass_check_output
if not pass_check_output:
print 'did not pass output check, bailiing before merging the files'
return False
print 'passed output check... proceding to file merging'
# next, merge the the files in output directory
RunHelpers.mergeOutputFiles(output_dir, flat_ntuples)
def main():
is_data = sys.argv[1]
is_full_sim = sys.argv[2]
input_list = sys.argv[3]
print 'is_data: %s' % is_data
print 'is_full_sim: %s' % is_full_sim
print 'input_list: %s' % input_list
file_list = RunHelpers.getFileListFromGridInput(input_list)
print 'file_list: %s' % file_list
runTntMaker( file_list
, is_data = is_data
, is_full_sim = is_full_sim
)
def runBMinusLAnalysis( file_list
, is_data
, is_egamma_stream
, is_full_sim
, syst_struct = None
, tree_name = 'susy'
, dsid = 1
, out_file_special_name = None
, is_tnt = False
, fancy_progress_bar = True
, job_num = 0
, total_num_jobs = 1
, total_num_events = 0
, total_num_entries = 0
, sum_mc_event_weights = 0
, out_dir = './'
):
# ==============================================================================
# If the num events are not set and we are running over TNTs, get the total
# NumEvents
print 'total num events: ', total_num_events
if total_num_events == 0 and is_tnt:
print 'Getting total num unskimmed events'
print ' -- this is slow. you should do this once per data set - not for each stream!'
total_num_events = getTotalNumEvents(file_list, is_tnt)
print "Adding files to TChain"
print ' Tree name: ' , tree_name
t = RunHelpers.getTChain(file_list, tree_name)
print t
print t.GetEntries()
# ==============================================================================
print 'Creating BMinusLAnalysis object'
bmla = ROOT.PennSusyFrame.BMinusLAnalysis(t)
print 'configuring BMinusLAnalysis object'
if out_file_special_name is not None:
process_label = ''.join( [ out_file_special_name
, '__'
, str(job_num)
, '_of_'
, str(total_num_jobs)
]
)
bmla.setProcessLabel(process_label)
bmla.setFancyProgressBar(False)
# set is data or MC
# if MC, we need to set various other things like cross section, k-factor, ...
if is_data:
bmla.setIsData()
if is_egamma_stream:
bmla.setIsEgammaStream()
else:
bmla.setIsMuonStream()
else:
bmla.setIsMC()
xsec_dict = CrossSectionReader.getCrossSection(dsid)
if xsec_dict is None:
return
bmla.setCrossSection(xsec_dict['xsec'])
bmla.setKFactor( xsec_dict['kfac'])
bmla.setFilterEff( xsec_dict['eff'])
bmla.setTotalNumEntries( total_num_entries )
bmla.setNumGeneratedEvents( total_num_events )
bmla.setSumMCEventWeights( sum_mc_event_weights )
# set is full sim/fast sim
if is_full_sim:
bmla.setFullSim()
# set unblind!
bmla.setIsBlind(False)
# turn on systematics
syst_tag = ''
if syst_struct:
print 'turning on systematics'
syst_struct.configureAnalysisObject(bmla)
syst_tag = ''.join([syst_struct.getRunName(), '.'])
# set start entry and max number events
if total_num_jobs > 1:
print 'total num jobs (', total_num_jobs, ') > 1'
this_job_events = int( math.ceil( float(total_num_entries)
/ total_num_jobs
)
)
this_job_start = job_num*this_job_events
print 'total num entries: ', total_num_entries
print 'setting max num events: ', this_job_events
print type(this_job_events)
bmla.setMaxNumEvents(this_job_events)
print 'setting start entry: ', this_job_start
bmla.setStartEntry(this_job_start)
# set out histogram file name
# base name
out_ntup_file_name = [out_dir, '/', 'BMinusL.', syst_tag]
# append any special tags
if out_file_special_name is not None:
out_ntup_file_name.extend([out_file_special_name, '.'])
# append 'hist' tag
out_ntup_file_name.append('hist')
# append job number
if total_num_jobs > 1:
out_ntup_file_name.extend( ['.'
, str(job_num)
, '_of_'
, str(total_num_jobs)
]
)
# this is a root file
out_ntup_file_name.append('.root')
# set output file name in analyzer
bmla.setOutHistFileName(''.join(out_ntup_file_name))
# Set critical cuts
bmla.setCritCutGrl( 1)
bmla.setCritCutIncompleteEvent(1)
bmla.setCritCutLarError( 1)
bmla.setCritCutTileError( 1)
bmla.setCritCutTileHotSpot( 1)
bmla.setCritCutTileTrip( 1)
bmla.setCritCutBadJetVeto( 1)
bmla.setCritCutCaloProblemJet( 1)
bmla.setCritCutPrimaryVertex( 1)
bmla.setCritCutBadMuonVeto( 1)
bmla.setCritCutCosmicMuonVeto( 1)
bmla.setCritCutHFOR( 1)
bmla.setCritCutMcOverlap( 1)
bmla.setCritCutGe2Lepton( 1)
bmla.setCritCut2Lepton( 0)
bmla.setCritCut2SignalLepton( 0)
bmla.setCritCut2BJets( 0)
bmla.setCritCutBadJetVeto( 1)
bmla.setCritCutBLPairing( 0)
# Set cut values
bmla.setElPtCut( lep_pt_cut, -1 )
bmla.setMuPtCut( lep_pt_cut, -1 )
bmla.setBJetPtCut(jet_pt_cut, -1 )
bmla.setMV1Cut(btag_working_point)
# Turn on/off Z fudge factor
bmla.setDoZKFactor(True)
# bmla.setDoZKFactor(False)
# Turn on/off detailed B-L histograms
bmla.setDoDetailedBLHists(False)
# prepare tools and run analysis loop
print 'preparing tools'
bmla.prepareTools()
print 'looping -- ', out_file_special_name
bmla.Loop()
print 'done looping -- ', out_file_special_name
# ==============================================================================
print ''
print ''
systematic_runs = itertools.chain([None], RunHelpers.syst_list)
for syst in systematic_runs:
syst_struct = RunHelpers.SystematicStruct()
if syst is not None:
syst_struct.setSyst(syst, True)
syst_tag = syst_struct.getRunName()
syst_struct.printInfo()
print syst_tag
data_set_dicts[syst_tag] = RunHelpers.addAllSamplesToList(
egamma_data_samples = egamma_data_samples,
muon_data_samples = muon_data_samples,
full_sim_mc_samples = full_sim_mc_samples,
fast_sim_mc_samples = fast_sim_mc_samples,
file_list_path_base = 'EosFileLists/tnt_107/tnt_107',
out_dir = '__'.join([out_dir, syst_tag]),
syst_struct = syst_struct)
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
for syst, the_dicts in data_set_dicts.items():
print syst, ' -- ', the_dicts
this_out_dir = '__'.join([out_dir, syst])
this_sym_link_name = ''.join(['./NextPlotDir.BMinusL.', syst])
print 'this sym link name: ', this_sym_link_name
this_run_analysis_fun = RunBMinusLAnalysis.runBMinusLAnalysisFun
if run_local:
def runIsrAnalysis( file_list
, is_data
, is_full_sim
, tree_name = 'susy'
, dsid = 1
, out_file_special_name = None
, is_tnt = False
, fancy_progress_bar = True
, job_num = 0
, total_num_jobs = 1
, total_num_events = 0
, total_num_entries = 0
, out_dir = './'
):
# ==============================================================================
# If the num events are not set and we are running over TNTs, get the total NumEvents
print 'total num events: %s' % total_num_events
if total_num_events == 0 and is_tnt:
print 'Getting total num unskimmed events'
print ' -- this is slow. you should do this once per data set - not for each stream!'
total_num_events = getTotalNumEvents(file_list, is_tnt)
print "Adding files to TChain"
t = RunHelpers.getTChain(file_list, tree_name)
# ==============================================================================
print 'Creating IsrAnalysis object'
isra = ROOT.PennSusyFrame.IsrAnalysis(t)
print 'configuring IsrAnalysis object'
if out_file_special_name is not None:
isra.setProcessLabel(out_file_special_name)
isra.setFancyProgressBar(False)
# set is data or MC
if is_data:
isra.setIsData()
else:
isra.setIsMC()
xsec_dict = CrossSectionReader.getCrossSection(dsid)
if xsec_dict is None:
return
isra.setCrossSection(xsec_dict['xsec'])
isra.setKFactor( xsec_dict['kfac'])
isra.setFilterEff( xsec_dict['eff'])
isra.setTotalNumEntries( total_num_entries )
isra.setNumGeneratedEvents( total_num_events )
# set is full sim/fast sim
if is_full_sim:
isra.setFullSim()
# set start entry and max number events
if total_num_jobs > 1:
print 'total num jobs (%s) > 1' % total_num_jobs
this_job_events = int(math.ceil( float(total_num_entries) / total_num_jobs ))
this_job_start = job_num*this_job_events
print 'total num entries; %s' % total_num_entries
print 'setting max num events: %s' % this_job_events
print type(this_job_events)
isra.setMaxNumEvents(this_job_events)
print 'setting start entry: %s' % this_job_start
isra.setStartEntry(this_job_start)
# set out histogram file name
print 'setting histogram names'
out_hist_file_name = '%s/Isr.' % out_dir
if out_file_special_name is not None:
out_hist_file_name += '%s.' % out_file_special_name
out_hist_file_name += 'hists'
if total_num_jobs > 1:
out_hist_file_name += '.%d_of_%d' % (job_num, total_num_jobs)
out_hist_file_name += '.root'
isra.setOutHistFileName(out_hist_file_name)
# Set critical cuts
print 'setting critical cuts'
# isra.setCritCutGrl( 1)
# isra.setCritCutIncompleteEvent(1)
# isra.setCritCutLarError( 1)
# isra.setCritCutTileError( 1)
# isra.setCritCutTileHotSpot( 1)
# isra.setCritCutTileTrip( 1)
# isra.setCritCutBadJetVeto( 1)
# isra.setCritCutCaloProblemJet( 1)
# isra.setCritCutPrimaryVertex( 1)
# isra.setCritCutBadMuonVeto( 1)
# isra.setCritCutCosmicMuonVeto( 1)
# isra.setCritCutHFOR( 1)
# isra.setCritCutMcOverlap( 1)
# isra.setCritCutGe2Lepton( 1)
# isra.setCritCut2Lepton( 1)
# isra.setCritCut2SignalLepton( 1)
# prepare tools and run analysis loop
print 'preparing tools'
isra.prepareTools()
print 'looping -- %s' % out_file_special_name
isra.Loop()
print 'done looping -- %s' % out_file_special_name
# ==============================================================================
print ''
print ''
num_processes = int(user_input)
run_local = True
today_date = datetime.datetime.now()
out_dir = '%s/hists/bminusl_hists_%04d_%02d_%02d__%02d_%02d' % ( os.environ['PWD']
, today_date.year
, today_date.month
, today_date.day
, today_date.hour
, today_date.minute
)
print out_dir
# ==============================================================================
if __name__ == '__main__':
RunHelpers.safeMakeDir(out_dir)
print 'getting file list'
data_samples = {
}
full_sim_mc_samples = {
117050:{'label':'117050.PowhegPythia_P2011C_ttbar' , 'num_jobs':30}
, 105200:{'label':'105200.McAtNloJimmy_CT10_ttbar_LeptonFilter' , 'num_jobs':30}
, 200332:{'label':'200332.AlpgenPythia_Auto_P2011C_ZeebbNp0' , 'num_jobs':20}
, 200333:{'label':'200333.AlpgenPythia_Auto_P2011C_ZeebbNp1' , 'num_jobs':10}
, 200334:{'label':'200334.AlpgenPythia_Auto_P2011C_ZeebbNp2' , 'num_jobs':3}
, 200335:{'label':'200335.AlpgenPythia_Auto_P2011C_ZeebbNp3incl' , 'num_jobs':1}
, 200340:{'label':'200340.AlpgenPythia_Auto_P2011C_ZmumubbNp0' , 'num_jobs':20}
, 200341:{'label':'200341.AlpgenPythia_Auto_P2011C_ZmumubbNp1' , 'num_jobs':10}
, 200342:{'label':'200342.AlpgenPythia_Auto_P2011C_ZmumubbNp2' , 'num_jobs':3}
import RunTntMaker
sys.path.append('%s/RunHelpers/' % os.environ['BASE_WORK_DIR'])
import RunHelpers
# ==============================================================================
if __name__ == '__main__':
base_work_dir = os.environ['BASE_WORK_DIR']
print 'getting file list'
data_set_dicts = []
RunHelpers.addSamplesToList( sample_dict = {'label':'202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000' , 'num_jobs':1}
, data_set_dicts = data_set_dicts
# , file_list_path = '%s/EosFileLists/local/raw_d3pd.202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000.txt' % os.environ['BASE_WORK_DIR']
, file_list_path = 'EosFileLists/local/raw_d3pd.202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000.txt'
, is_data = False
, is_egamma_stream = False
, is_full_sim = False
, dsid = 202641
, out_dir = 'local_tnt'
)
print ''
print data_set_dicts
print ''
RunTntMaker.runTntMaker( file_list = data_set_dicts[0]['file_list']
, is_data = False
, is_full_sim = False
)
# # file_list = RunTntMaker.getFileListFromFile('%s/EosFileLists/local/raw_d3pd.202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000.txt' % os.environ['BASE_WORK_DIR'])
# # file_list = file_list[:1]
from multiprocessing import Pool
import time
import sys
import datetime
sys.path.append('%s/EwkAnalysis/RunScripts/' % os.environ['BASE_WORK_DIR'])
import RunIsrAnalysis
sys.path.append('%s/RunHelpers/' % os.environ['BASE_WORK_DIR'])
import RunHelpers
# ------------------------------------------------------------------------------
# get number of parallel processes from command line inputs
num_processes = int(sys.argv[1]) if len(sys.argv) > 1 else 1
out_dir = 'hists/isr_hists%s' % RunHelpers.getDateTimeLabel(True, False)
# ==============================================================================
if __name__ == '__main__':
RunHelpers.safeMakeDir(out_dir)
print 'getting file list'
data_set_dicts = []
data_samples = {}
full_sim_mc_samples = {## 117650:{'label':'117650.AlpgenPythia_P2011C_ZeeNp0' , 'num_jobs':10}
## , 117651:{'label':'117651.AlpgenPythia_P2011C_ZeeNp1' , 'num_jobs':10}
## , 117652:{'label':'117652.AlpgenPythia_P2011C_ZeeNp2' , 'num_jobs':5}
## , 117653:{'label':'117653.AlpgenPythia_P2011C_ZeeNp3' , 'num_jobs':1}
## , 117654:{'label':'117654.AlpgenPythia_P2011C_ZeeNp4' , 'num_jobs':1}
## , 117655:{'label':'117655.AlpgenPythia_P2011C_ZeeNp5' , 'num_jobs':1}
run_local = True
today_date = datetime.datetime.now()
out_dir = '%s/hists/bminusl_opt_ntup_%04d_%02d_%02d__%02d_%02d' % ( os.environ['PWD']
, today_date.year
, today_date.month
, today_date.day
, today_date.hour
, today_date.minute
)
print out_dir
# ==============================================================================
if __name__ == '__main__':
RunHelpers.safeMakeDir(out_dir)
print 'getting file list'
egamma_data_samples = {
'periodA_egamma':{'label':'periodA_egamma', 'num_jobs':2}
, 'periodB_egamma':{'label':'periodB_egamma', 'num_jobs':2}
, 'periodC_egamma':{'label':'periodC_egamma', 'num_jobs':2}
, 'periodD_egamma':{'label':'periodD_egamma', 'num_jobs':2}
, 'periodE_egamma':{'label':'periodE_egamma', 'num_jobs':2}
, 'periodG_egamma':{'label':'periodG_egamma', 'num_jobs':2}
, 'periodH_egamma':{'label':'periodH_egamma', 'num_jobs':2}
, 'periodI_egamma':{'label':'periodI_egamma', 'num_jobs':2}
, 'periodJ_egamma':{'label':'periodJ_egamma', 'num_jobs':2}
, 'periodL_egamma':{'label':'periodL_egamma', 'num_jobs':2}
}
# ------------------------------------------------------------------------------
# get number of parallel processes from command line inputs
num_processes = int(sys.argv[1]) if len(sys.argv) > 1 else 1
today_date = datetime.datetime.now()
out_dir = 'hists/bminusl_hists_%04d_%02d_%02d__%02d_%02d' % ( today_date.year
, today_date.month
, today_date.day
, today_date.hour
, today_date.minute
)
print out_dir
# ==============================================================================
if __name__ == '__main__':
RunHelpers.safeMakeDir(out_dir)
print 'getting file list'
data_set_dicts = []
data_samples = {
}
full_sim_mc_samples = {
}
fast_sim_mc_samples = {
202641:{'label':'202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000' , 'num_jobs':1}
}
print fast_sim_mc_samples
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def runBMinusLTTNTMaker( file_list
, is_data
, is_egamma_stream
, is_full_sim
, tree_name = 'susy'
, dsid = 1
, out_file_special_name = None
, is_tnt = False
, fancy_progress_bar = True
, job_num = 0
, total_num_jobs = 1
, total_num_events = 0
, total_num_entries = 0
, sum_mc_event_weights = 0
, out_dir = './'
):
# ==============================================================================
# If the num events are not set and we are running over TNTs, get the total NumEvents
print 'total num events: %s' % total_num_events
if total_num_events == 0 and is_tnt:
print 'Getting total num unskimmed events'
print ' -- this is slow. you should do this once per data set - not for each stream!'
total_num_events = getTotalNumEvents(file_list, is_tnt)
print "Adding files to TChain"
t = RunHelpers.getTChain(file_list, tree_name)
# ==============================================================================
print 'Creating BMinusLTTNTMaker object'
bmlttntm = ROOT.PennSusyFrame.BMinusLTTNTMaker(t)
print 'configuring BMinusLTTNTMaker object'
if out_file_special_name is not None:
bmlttntm.setProcessLabel(out_file_special_name)
bmlttntm.setFancyProgressBar(False)
# set is data or MC
if is_data:
bmlttntm.setIsData()
else:
bmlttntm.setIsMC()
xsec_dict = CrossSectionReader.getCrossSection(dsid)
if xsec_dict is None:
return
bmlttntm.setCrossSection(xsec_dict['xsec'])
bmlttntm.setKFactor( xsec_dict['kfac'])
bmlttntm.setFilterEff( xsec_dict['eff'])
print 'Setting TotalNumEntries: ' , total_num_entries
print 'Setting NumGeneratedEvents: ' , total_num_events
print 'Setting SumMCEventWeights: ' , sum_mc_event_weights
bmlttntm.setTotalNumEntries( total_num_entries )
bmlttntm.setNumGeneratedEvents( total_num_events )
bmlttntm.setSumMCEventWeights( sum_mc_event_weights )
# set is full sim/fast sim
if is_full_sim:
bmlttntm.setFullSim()
# set start entry and max number events
if total_num_jobs > 1:
print 'total num jobs (%s) > 1' % total_num_jobs
this_job_events = int(math.ceil( float(total_num_entries) / total_num_jobs ))
this_job_start = job_num*this_job_events
print 'total num entries; %s' % total_num_entries
print 'setting max num events: %s' % this_job_events
print type(this_job_events)
bmlttntm.setMaxNumEvents(this_job_events)
print 'setting start entry: %s' % this_job_start
bmlttntm.setStartEntry(this_job_start)
# set out histogram file name
print 'setting histogram names'
out_ntup_file_name = '%s/BMinusL.' % out_dir
if out_file_special_name is not None:
out_ntup_file_name += '%s.' % out_file_special_name
out_ntup_file_name += 'tnt'
if total_num_jobs > 1:
out_ntup_file_name += '.%d_of_%d' % (job_num, total_num_jobs)
out_ntup_file_name += '.root'
bmlttntm.setOutNtupleFileName(out_ntup_file_name)
# # Set critical cuts
# print 'setting critical cuts'
# bmlttntm.setCritCutGrl( 1)
# bmlttntm.setCritCutIncompleteEvent(1)
# bmlttntm.setCritCutLarError( 1)
# bmlttntm.setCritCutTileError( 1)
# bmlttntm.setCritCutTileHotSpot( 1)
# bmlttntm.setCritCutTileTrip( 1)
# bmlttntm.setCritCutBadJetVeto( 1)
# bmlttntm.setCritCutCaloProblemJet( 1)
# bmlttntm.setCritCutPrimaryVertex( 1)
# bmlttntm.setCritCutBadMuonVeto( 1)
# bmlttntm.setCritCutCosmicMuonVeto( 1)
# bmlttntm.setCritCutHFOR( 1)
# bmlttntm.setCritCutMcOverlap( 1)
# bmlttntm.setCritCutGe2Lepton( 1)
# bmlttntm.setCritCut2Lepton( 0)
# bmlttntm.setCritCut2SignalLepton( 0)
# bmlttntm.setCritCut2BJets( 0)
# bmlttntm.setCritCutBadJetVeto( 1)
# bmlttntm.setCritCutBLPairing( 0)
# Set cut values
bmlttntm.setElPtCut( lep_pt_cut, -1 )
bmlttntm.setMuPtCut( lep_pt_cut, -1 )
bmlttntm.setBJetPtCut(jet_pt_cut, -1 )
bmlttntm.setMV1Cut(btag_working_point)
# prepare tools and run analysis loop
print 'preparing tools'
bmlttntm.prepareTools()
print 'looping -- %s' % out_file_special_name
bmlttntm.Loop()
print 'done looping -- %s' % out_file_special_name
# ==============================================================================
print ''
print ''
# ------------------------------------------------------------------------------
# get number of parallel processes from command line inputs
num_processes = int(sys.argv[1]) if len(sys.argv) > 1 else 1
today_date = datetime.datetime.now()
out_dir = 'hists/bminusl_hists_%04d_%02d_%02d__%02d_%02d' % ( today_date.year
, today_date.month
, today_date.day
, today_date.hour
, today_date.minute
)
print out_dir
# ==============================================================================
if __name__ == '__main__':
RunHelpers.safeMakeDir(out_dir)
print 'getting file list'
data_set_dicts = []
data_samples = {
}
# full_sim_mc_samples = { # 110820:{'label':'110820.AlpgenPythia_P2011C_ZeebbNp3' , 'num_jobs':1}
# # 200334:{'label':'200334.AlpgenPythia_Auto_P2011C_ZeebbNp2' , 'num_jobs':3}
# }
# fast_sim_mc_samples = { 202641:{'label':'202641.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_1000', 'num_jobs':1}
# }
full_sim_mc_samples = { }
fast_sim_mc_samples = { # 105200:{'label':'105200.McAtNloJimmy_CT10_ttbar_LeptonFilter', 'num_jobs':15}
202632:{'label':'202632.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_100' , 'num_jobs':1}
, 202633:{'label':'202633.MadGraphPythia_AUET2B_CTEQ6L1_SM_TT_directBL_200' , 'num_jobs':1}
