package to run different analysis configurations with the same class. Common classes defined in the package
Handler.py: main class to define RDataFrame object reading all the root files stored in a given directory; the following methods are defined:cut: cut="preselection string" defined a preselection to be applied to all the events.AddLumiWeight(LumiTarget="1.",xsecName=""): add a luminosity weight (labelled "lumiWeight") to the main tree.Histo1D(hName,var,weight,Nbins,Xmin,Xmax,cut="true"): define a 1D histogram after having specified the weight to be applied (weight), interval and numbers of bins, additional cut to be applied.Histo2D(hName,var,weight,NbinsX,Xmin,Xmax,NbinsY,Ymin,Ymax,cut="true"): same asHisto1Dbut for a 2D histo.WriteHisto(outFileName,fMode="RECREATE"): write crated histos in a output root file.WriteTree(outTreeName,outFileName,columns=None,fMode="RECREATE"): write a tree with the leaves specified incolumns(ifNoneis specified the same leaves of input tree plus the ones defined by the user selection are stored) in a new root file. Be careful that in this method there is a hack to let RDataFrame store the std::vector objects in the output file; if needed modified this method.
XSecDataBase: class storing a dictionary of the cross-section of the various samples used in the analysis.
Example of a user class:
Analysis_Ak8Tagger.py: child class ofHandler.pyused to skim the initial tree, to add new variables and to produce 1D/2D histograms for AK8Analysis.
The code has been tested with python3 and root 6.26. To run the code just simply source setup.sh.
Under scripts there is the basic example submitAna_example.py to process a MC dataset with Analysis_Ak8Tagger.py.
submitAna_example.py -i /lustre/cms/store/user/azaza/QCD_PT-600to800_TuneCP5_13p6TeV_pythia8/crab_QCD_PT-600to800_TuneCP5_13p6TeV_pythia8_Run3Summer22MiniAODv3-124X_mcRun3_2022_realistic_v12-v2/230405_184254/ -L 8924. -s QCD_PT-600to800_TuneCP5_13p6TeV_pythia8 -N 100 |& tee output_test
INFO:QCD_PT-600to800_TuneCP5_13p6TeV_pythia8:Reading files: /lustre/cms/store/user/azaza/QCD_PT-600to800_TuneCP5_13p6TeV_pythia8/crab_QCD_PT-600to800_TuneCP5_13p6TeV_pythia8_Run3Summer22MiniAODv3-124X_mcRun3_2022_realistic_v12-v2/230405_184254/
INFO:QCD_PT-600to800_TuneCP5_13p6TeV_pythia8:- LumiTarget=8924.
INFO:QCD_PT-600to800_TuneCP5_13p6TeV_pythia8:- TotEvents=7840148.0
INFO:QCD_PT-600to800_TuneCP5_13p6TeV_pythia8:- xsecDB=180.1
INFO:QCD_PT-600to800_TuneCP5_13p6TeV_pythia8:- lumiweight=0.2049977117778899
Preselection: pass=100 all=100 -- eff=100.00 % cumulative eff=100.00 %
TriggerMatch: pass=100 all=100 -- eff=100.00 % cumulative eff=100.00 %
LeptonVeto: pass=100 all=100 -- eff=100.00 % cumulative eff=100.00 %
AK8nJet>2 : pass=100 all=100 -- eff=100.00 % cumulative eff=100.00 %
AK8Jet0 (pt>450 GeV && eta<2.4): pass=99 all=100 -- eff=99.00 % cumulative eff=99.00 %
AK8Jet1 (pt>200 GeV && eta<2.4): pass=98 all=99 -- eff=98.99 % cumulative eff=98.00 %
All outputs produced in 0.0h:0.0m:26.9s
The code can be run in multithread with the option -MT (this option can not be run together with -N). More option are available in submit_example.py:
parser.add_argument("-l", "--logLevel", help="set log level", default="INFO", choices=["DEBUG", "INFO"])
parser.add_argument("-MT","--MultiThread", help="enable multithread", action="store_true")
parser.add_argument("-i","--InputPath", help="Input root files path",default="")
parser.add_argument("-t","--TreeName", help="name of the tree to be processed, i.e. Ana/passedEvents",default="Ana/passedEvents")
parser.add_argument("-H","--HistoName", help="name of the histo stoting the total number of events, i.e. Ana/nEvents",default="Ana/nEvents")
parser.add_argument("-ot","--OutputTreeName", help="tree name to be used for the output",default="passedEvents")
parser.add_argument("-otf","--OutputFileTreeName", help="file name of the output file to store trees",default="Output_tree.root")
parser.add_argument("-ohf","--OutputFileHistoName", help="file name of the output file to store histos",default="Output_histo.root")
parser.add_argument("-p","--PreselectionCut", help="preselection cut to be applied",default="")
parser.add_argument("-L","--Lumi", help="luminosity to target in pb",default="-1")
parser.add_argument("-s","--SampleName", help="sample name used in the xsec database",default="sample")
parser.add_argument("-N","--NEvents", help="select NEvents to be processed", default="-1")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent