L1T Jet layer 2 calibration

Script to make L1T JEC LUTs

Step 1: Produce L1T Nano Files

• Recipe to make L1T Nanos can be found at here. Choose the latest CMSSW release and l1t-integration tag.

• Dummy cmsDriver command for running over Muon1/RAW-RECO(change the input file if not available). Make sure you have required caloparams (L1TSettingsToCaloParams_2025_baseline_v2_iET) correctly defined in L1Trigger/Configuration/python/customiseSettings.py, if not, refer here for instructions.

    cmsDriver.py customL1toNANO --conditions 140X_dataRun3_v20 -s RAW2DIGI,L1Reco,NANO:@PHYS+@L1DPG --datatier NANOAOD --eventcontent NANOAOD --data --process customl1nano --scenario pp --era Run3 --customise Configuration/DataProcessing/RecoTLR.customisePostEra_Run3 --customise_unsch L1Trigger/Configuration/customiseReEmul.L1TReEmulFromRAW --customise_unsch L1Trigger/Configuration/customiseSettings.L1TSettingsToCaloParams_2025_baseline_v2_iET --secondfilein /store/data/Run2024I/Muon0/RAW/v1/000/386/505/00000/46111224-ec19-4680-b777-4a175e09659e.root --filein /store/data/Run2024I/Muon1/RAW-RECO/ZMu-PromptReco-v1/000/386/679/00000/429dca6f-1658-46fe-a9f1-b45c16530332.root -n -1 --no_exec
  

• Sample files are available in sample_l1nano_script/ with sample crab submit file.

Step 2: Produce BDT input .csv file

The produced L1Nano files are read and BDT input in .csv files are produced by BDTInputProducer/script.py.

Input parameters read by default:

• --l1nano: ../sample_root_files/Nano.root.root
• --output: Nano_out
• --l1MatchOffline: True
• --l1nanoPhiRing

Step 3: BDT trainning

1. Calibration JEC SF computation script is setup/calculate_L1JetSFs_usingBD.ipynb.

2. Run for 4 different setups:

   python3 calculate_L1JetSFs_usingBDT.py --PhiRing     --l1MatchGen --MLTarget GenEt
   
   python3 calculate_L1JetSFs_usingBDT.py --PhiRing     --l1MatchGen --MLTarget logGenEt
   
   python3 calculate_L1JetSFs_usingBDT.py --PhiRing     --l1MatchGen --MLTarget GenEtByL1Et
   
   python3 calculate_L1JetSFs_usingBDT.py --PhiRing     --l1MatchGen --MLTarget logGenEtByL1Et
   

   Modify --fracOfDataToUse accordingly.

3. Run setup/compare_JEC_SFs.ipynb to get final JEC_SF.csv file.

4. Run setup/check_L1JetSFs.ipynb to make quality control plots.

Note: Outputs will be written in the data/ directory.

Step 4: Make LUTs

1. cd makeLUTs

2. Run python3 updateSFPtEtaBins.py

3. Run check_L1TJetSFs.ipynb to make performance plots using the new JEC LUTs

git commit:

git remote add origin [email protected]:dharmendergaur/JEC.git git push -u origin master