Hmumu

Repository for the Hmumu analysis

General Information

Central NANOAOD production, taken from Nano AOD documentation

How to install the FW

1. Setup ssh keys: On GitHub settings/keys On CERN GitLab profile/keys
2. git clone --recursive git@github.com:cms-flaf/H_mumu.git
3. (optional but recommended) create a fork from the central repository: 3.1. Go on central repository 3.2. Follow the instructions from github documentation for creating a fork 3.3. In your work-area (or where you cloned the repository - workarea recommended) in the H_mumu cloned folder, follow these instructions
4. Create a file in the folder: H_mumu/config/user_custom.yaml The content of the file should be: - currently it's fine but it will be changed

   fs_default:
   - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/'
   fs_anaCache:
       - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/'
   fs_anaTuple:
       - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/'
   fs_anaTuple2p5:
       - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/'
   fs_nnCacheTuple:
       - 'T3_CH_CERNBOX:/store/user/prsolank/H_mumu/'
   fs_anaCacheTuple:
       - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/'
   fs_histograms:
       - 'T3_CH_CERNBOX:/store/user/vdamante/H_mumu/histograms/'
   
   analysis_config_area: config
   
   compute_unc_variations: False
   store_noncentral: False
   compute_unc_histograms: False
   
   variables:
       - mu1_pt
   

Setup steps

1. Load environment: source env.sh

2. Add law tasks to the index: law index (if you want to know more about what is happening, you can add --verbose)

3. Initialize voms-proxy voms-proxy-init -voms cms -rfc -valid 192:00 (if you do not have grid permission and certificate, see this twiki)

Get input files

This step consists in saving filenames in local (data/InputFileTask/...) and is required for all subsequent steps: law run InputFileTask --period ${ERA} --version ${VERSION_NAME} where you can either set the ${ERA} or just manually put the era name (Run3_2022,Run3_2022EE etc) and version consists in a general name (e.g. Run3_2022)

Get AnaCache

The anaCaches are yaml file storing inputs from both saved and non-saved events from original NanoAOD (needed for pu uncertainty)

law run AnaCacheTask --period ${ERA} --version ${VERSION_NAME}

The ${ERA} is defined as before, whereas version can differ from InputFileTask (e.g. AnaCacheVersion is Run3_2022_v1 and InputFileTaskVersion is Run3_2022). The important is that if they differ you need also to add

 --InputFileTask-version ${INPUTFILETASK_VERSION_NAME}

which in this example is Run3_2022.

ToDo list

• redesign jet selection and store
• modify H(mm) candate selection
• check what is done for MET in Run2 analysis
• check the b-tagging WP based SF for jets
• include FSR
• add all corrections
• data/MC control plots (short term)
• setup workspace for ggH and VBF (for the moment both template fit)
• improve documentations + add github project to correctly keep trace of changes

R&D

• efficiency gain using other triggers (non priority)
• develop classifier optimized for Run3 (first focus on VBF, then possible extension to all categories)
  • explore different architechtures
  • validate
  • hyperparameter optimization
  • suitable to compare results w/ run2 and for run2+run3 combination
• switch to Run3 muon ID
• investigate for new production setup (e.g. flashsim)