• Pushpa, Oct 17
  • There seems to be jet flavor tag/prob available to use with miniAOD for 9_4_5 and later pfDeepFlavourJetTags:probxxx, where xxx is b, bb, lepb, c, uds, g
  • dataset dataset=/*/RunIIFall17*/AODSIM
  • Deep-Flavor wiiki https://twiki.cern.ch/twiki/bin/view/CMS/DeepJet
• Nicolas code for HZZ
  • https://github.com/ndefilip
• MiniAOD workbook https://twiki.cern.ch/twiki/bin/view/CMSPublic/WorkBookMiniAOD

Working area at cms1

• ssh cms1
  • cd /cdata/serguei/; cmset; cmsrel CMSSW_9_4_9
  • cd CMSSW_9_4_9; smsenv

Now checkout/build package for https://github.com/CJLST/ZZAnalysis

• cd tmp/
• wget https://raw.githubusercontent.com/CJLST/ZZAnalysis/miniAOD_80X/checkout_9X.csh
• chmod 755 checkout_9X.csh
• cd src; ../tmp/checkout_9X.csh
• cd ../; time scram build -j 12
  • compilation is ok real 759.485 user 5927.475 sys 1196.139 pcpu 938.05

Getting sample to work with

• Open voms session ( on a desktop browser need have grid certificate loaded)
  • voms-proxy-init -voms cms
• Open https://cmsweb.cern.ch/das
• query example dataset release=CMSSW_9_4* datatype=mc
• get a file from ttHiggs0PMToGG_M125_13TeV_JHUGenV7011_pythia8 dataset
  • xrdcp root://cms-xrd-global.cern.ch//store/mc/RunIIFall17MiniAODv2/ttHiggs0PMToGG_M125_13TeV_JHUGenV7011_pythia8/MINIAODSIM/PU2017_12Apr2018_94X_mc2017_realistic_v14-v1/00000/0E0FAF14-6183-E811-BC58-FA163EADDDE5.root ./
    • [2.586GB/2.586GB][100%]================================================[46.45MB/s]
  • mv 0E0FAF14-6183-E811-BC58-FA163EADDDE5.root ttHiggs0PMToGG_M125_13TeV_JHUGenV7011_pythia8_miniAOD.root

Reham's suggestion

Dear All,

I added some information about b tagging in our code for 2017, I think it is part of what you need

In file hTozzTo4leptonsCommonRootTree_cfi.py, I used pfDeepCSVJetTags:probb & pfDeepCSVJetTags:probbb in the following lines:

    cSV_bTagLabel1        = cms.string("pfDeepCSVJetTags:probb"),#2017
    cSV_bTagLabel2        = cms.string("pfDeepCSVJetTags:probbb"),#2017

Then in file HZZ4LeptonsCommonRootTree.h I added the value of the two discriminator in lines

    double discrCSV3_1 = btagIter->bDiscriminator(cSV_bTag1_);
    double discrCSV3_2 = btagIter->bDiscriminator(cSV_bTag2_);
    double discrCSV3 = discrCSV3_1 + discrCSV3_2 ;

The value of the working point for b tagging discrimentor => medium working point = 0.4941

Regards Reham

1. ==================

on behalf of Nicola De Filippis Sent: 17 October 2018 23:45 Subject: Re: jet flavor tags

that can be done with the 2017 analysis and code…before running on all the samples of data and MC

Nic

@]