Skip to content

Scale factors from JSONs

The following is the guide to use EGamma JSON's for getting electron/photon scale factors in your analysis:

First Step: Get the correctionlib library

python3 -m pip install git+https://github.com/cms-nanoAOD/correctionlib.git

Second Step: Get the EGamma JSONs

You can either get them from the central POG repository or get them from CVMFS: /cvmfs/cms.cern.ch/rsync/cms-nanoAOD/jsonpog-integration (recommended). The cvmfs files are synced once a day.

Electron JSON

This JSON has 1 format

Electron JSON Format:

evaluator[ JSON Map Name %ENDCOLOR%].evaluate( year , Value Type , Working Point , eta , pt )

Format Guide:
JSON to use JSON Map Name year Value Type Working Point eta pt
electron.json "UL-Electron-ID-SF" "2016preVFP", "2016postVFP", "2017","2018" "sf", "sfup","sfdown" "RecoBelow20","RecoAbove20","Veto", "Loose", "Medium", "Tight", "wp80iso", "wp80noiso", "wp90iso", "wp90noiso" float value float value
Python Example for electron JSON:

Python Example to print a scale factor value

from correctionlib import _core

#Download the correct JSON files 
evaluator = _core.CorrectionSet.from_file('electron.json')

#Reconstruction (pT< 20 GeV) Run-2 scale factor
valsf= evaluator["UL-Electron-ID-SF"].evaluate("2016postVFP","sf","RecoBelow20",1.1, 15.0)
print("sf is:"+str(valsf))

#Reconstruction (pT> 20 GeV) Run-2 scale factor
valsf= evaluator["UL-Electron-ID-SF"].evaluate("2016postVFP","sf","RecoAbove20",1.1, 25.0)
print("sf is:"+str(valsf))

Photon JSON

This JSON has 2 formats

Photon JSON Format 1 for ID SFs:

evaluator[ JSON Map Name %ENDCOLOR%].evaluate( year , Value Type , Working Point , eta , pt )

Format Guide:
JSON to use JSON Map Name year Value Type Working Point eta pt
photon.json "UL-Photon-ID-SF" "2016preVFP", "2016postVFP", "2017","2018" "sf", "sfup","sfdown" "Loose", "Medium", "Tight", "wp80", "wp90" float value float value
Photon JSON Format 2 for CSEV AND Pix veto SFs:

evaluator[ JSON Map Name %ENDCOLOR%].evaluate( year , Value Type , Working Point , Bin )

Format Guide:
JSON to use JSON Map Name year Value Type Working Point Bin
photon.json "UL-Photon-PixVeto-SF", "UL-Photon-CSEV-SF" "2016preVFP", "2016postVFP", "2017","2018" "sf", "sfup","sfdown" "Loose", "Medium", "Tight", "MVA" "EBInc","EBHighR9", "EBLowR9", "EEInc", "EEHighR9", "EELowR9"
Python Example for photon JSON:

Python Example to print a scale factor value for JSON Format 1

from correctionlib import _core

#Download the correct JSON files 
from evaluator = _core.CorrectionSet.from_file('photon.json')

valsyst= evaluator["UL-Photon-ID-SF"].evaluate("2016postVFP","sfup","Medium",1.1, 34.0)
print("sfup is:"+str(valsyst))

Python Example to print a scale factor value for JSON Format 2

from correctionlib import _core

#Download the correct JSON files 
from evaluator = _core.CorrectionSet.from_file('photon.json')

valsf= evaluator["UL-Photon-CSEV-SF"].evaluate("2016postVFP","sf","Loose","EBInc")
print("sf is:"+str(valsf))
The merging with common CMS repo can be monitored here:

https://gitlab.cern.ch/cms-nanoAOD/jsonpog-integration/-/merge_requests/37