Recipe for Latino Framework on SLHC samples

How to Install

 cmsrel CMSSW_6_2_0_SLHC20
 cd CMSSW_6_2_0_SLHC20/src 
 cmsenv
 cp -r /afs/cern.ch/user/r/rgerosa/public/xTP/CMSSW_6_2_0_SLHC20/src/RecoEgamma ./
 cp -r /afs/cern.ch/user/r/rgerosa/public/xTP/CMSSW_6_2_0_SLHC20/src/Dummy ./
 git clone https://github.com/latinos/LatinoTrees.git
 cd LatinoTrees 
 git checkout latino_SLHC_miniAOD
 cd ..
 scramv1 b -j 

Code Workflow

 cd LatinoTrees/AnalysisStep/test

The code to be run is named: miniAOD_step.py.

The workflow is the following:

  • Import the correct geometry and global tag that are related to the specific sample on which you are running (CMS, SHCal, HGCal .. etc)
  • Run the pat sequence to create standard pat electrons, photons, muon, jets and met
  • Run the same pat jet sequence on top of puppi particles
  • User latino analyzer to produce a final ROOT plain tree

Main options:

  • globalTag : in order to specify the right GT upgradeGT
  • cmsGeometry : in order to load the correct geometry for the analyzed sample upgradeGeom
  • selection : to specify a skim of the event, up to now no events are skimmed
  • doSameSign : to perform the same sign analysis
  • doNoFilter : to turn off any kind of event filter
  • doMETFilter : to turn on met filters (some of them are crashing on upgrade samples, by default set to false)
  • doLHE : to dump LHE file information
  • doGen : to dump Gen information
  • doAdditionalJets : to dump jets up to 8 for both puppi and standard collection
  • runPUPPISequence : to run puppi modules and save puppi jet information
  • producePATObjects : if true is used to stop the code at pattuple level storing pattuples instead of doing latino trees
  • jetIdWP : used to define the type of jetID to apply (1 means standard jet ID)
  • pileupjetIdWP : used to define type of pileup jet id to apply (0 no cut, 1 loose cut, 2 medium, 3 tight)
  • CJVminPt : minimum for looking at b-jets

Pat Sequence setup:

  • Electrons : electron id value map are created on top of gedGsfElectrons (RobustLoose,RobustTight,RobustHighEnergy,Loose,Tight), particle flow isolation values are calculated for dR=0,3,0.4, some information are embedded in the PAT object: recHits, core, PF candidate and clusters, electronID, GenMatch, isolation values.
  • Muons : particle flow isolation values are calculated for dR=0,3,0.4, some information are embedded in the PAT object: muonID, GenMatch, isolation values, muon best track, TuneP.
  • Photons : photon id value map are created on top of gedPhotons, particle flow isolation values are calculated for dR=0,3,0.4, some information are embedded in the PAT object: photonID, GenMatch, isolation values, recHits, core, PF candidate ..etc
  • Taus : removed from the sequence
  • Jets : default benchmark is AK5PFJets + CHS. A lot of info are stored inside pat jets: jet track association, jet charge, jet flavour, JEC (using fastjet grid for rho), gen jet matching, gen parton matching, jetID, pileUp jet id (flag value map and discriminant value), bDiscriminators (TCHE, TCHP, jetProbability, CSV, CSVMVA, SSVHE, SSVHP)
  • MET : to the Raw met, typeI correction for using AK5PFjets +CHS is applied (only jet with pt > 10 GeV used for TypeI correction)

Pat Object Selection:

  • Electrons : no cuts.
  • Muons : no cuts.
  • Photons : no cuts.
  • Jets : no cuts.

PUPPI Sequence:

  • Produce PUPPI particles and puppi patMET
  • Cluster PUPPI particles in AK5 jets
  • Run jet track association, gen jet and gen parton matching, jet flavour, pile-up jet id and b-tagging on top of PUPPI jets
  • Produce pat PUPPI jets that have the same content of the standard PAT jets
  • Here you can find the python code: puppiSequence

LATINO Sequence:

NEW VARIABLES:

  • Store the muon ID flag: isTightMuon, isLooseMuon, isHighPtMuon, isSoftMuon, isSTA
  • Store the electron ID flag: isTightElectron, isLooseElectron, isLooseRobustElectron, isTightRobustElectron,isRobustHighEnergyElectron
  • Store lepton PF ISO: pfNeutralHadronsIso, pfParticleAllIso, pfPUChargedHadronIso, pfChargedHadronsIso, pfPhotonsIso
  • Bveto flags added for CSV (loose, medium and tight WP)
  • Btag counter added for CSV (loose, medium and tight WP) for both cut based and MVA one.
  • Pileup jetID flag
  • Pileup jetID value
  • All jet infor are computed for both puppi and chs jets
  • Store up to 8 jets in the final tree without cutting in pt
  • Store up to 5 gen jets
  • Puppi MET

How to Run

example for a SCH file:

 cd LatinoTrees/AnalysisStep/test
 cmsRun miniAOD_stepB.py inputFiles=/store/relval/CMSSW_6_2_0_SLHC17/RelValZEE_14TeV/GEN-SIM-RECO/DES23_62_V1_UPG2023SHNoTaper-v1/00000/7466ABF1-652D-E411-A9CA-0025905964A6.root maxEvents=10 globalTag=DES23_62_V1::All cmsGeometry=Extended2023SHCalNoTaper runPUPPISequence=True

List of Exsisting Trees

Run on SLHC20 GEN-SIM-RECO samples. Multi Crab configuration file is here multicrab.cfg.

Output files in T2_CERN EOS:

 /store/caf/user/rgerosa/TPSAMPLES_14TEV/LATINO_TREE/RELVAL/
 

-- RaffaeleAngeloGerosa - 2014-11-05

Edit | Attach | Watch | Print version | History: r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r3 - 2014-11-06 - RaffaeleAngeloGerosa
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2021 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback