Skip to content

Code to estimate scale factors for W-taggers in a semileptonic ttbar sample based on a simultaneous fit of pass and fail samples.

Notifications You must be signed in to change notification settings

DAZSLE/boostedWScalefactorProducer

 
 

Repository files navigation

How to run the W-tagging scalefactor code

#########################################

installation instructions

Source ROOT version 5.34/36 independently (currently no CMSSW version with ROOT 5.34.X with X>18 available), e.g at PSI:

source /swshare/ROOT/root_v5.34.32_precompiled/root/bin/thisroot/.sh

getting the code

export GITUSER=`git config user.github`
echo "Your github username has been set to \"$GITUSER\""
git clone git@github.com:${GITUSER}/boostedWScalefactorProducer.git
cd boostedWScalefactorProducer
git remote add originalRemote git@github.com:thaarres/boostedWScalefactorProducer.git
git fetch originalRemote
git checkout -b DevelopmentBranch thaarres/DevelopmentBranch
cd $CMSSW_BASE/src
cd boostedWScalefactorProducer
export ROOFITSYS=$ROOTSYS
python Automatic_Setup.py --vclean 1

running

python Automatic_Setup.py #To compile
python wtagSFfits.py -b   #To run

The basic script to be run is

python wtagSFfits.py

It takes as input .root files containing a TTree with a branch for the mass distribution you want to calculate a scalefactor for. This branch can contain events after full selection is applied, or new selections can be implemented on the fly in wtagSFfits.py. In addition to a data and the separate background MC files, you need one file called "pseudodata wchich contains all MC added together (with their appropriate weights, using ROOT hadd).

General Options:

    -b : To run without X11 windows
    -c : channel you are using(electron,muon or electron+muon added together)
    --HP : HP working point
    --LP : LP working point
    --fitTT : Only do fits to truth matched tt MC
    --fitMC : Only do fits to MC (test fit functions)
    --sample : name of TT MC eg --sample "herwig"
    --doBinned : to do binned simultaneous fit (default is unbinned)
    --76X : Use files with postfix "_76X" (change to postfix of your choice if running on several different samples)
    --useDDT : Uses DDT tagger instead of pruning+softdrop (ops! Requires softdrop variables)
    --usePuppiSD : Uses PUPPI + softdrop and PUPPI n-subjettiness

About

Code to estimate scale factors for W-taggers in a semileptonic ttbar sample based on a simultaneous fit of pass and fail samples.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 54.5%
  • C++ 44.8%
  • Other 0.7%