Analysis nTuple for HMT trigger studies. This repository can do the following:
- Run HMT emulator
CSCShowerAnalyzer
writes emulator-check histograms
- Run CA4 clustering on HMT hits
- Save emulated/L1 LCT hits
- Run standard muon reco [experimental]
SimpleCSCshowerFilter
builds HMT tree to contain the above
cmsrel CMSSW_13_2_5
cd CMSSW_13_2_5/src
git clone git@github.com:kakwok/HMTntuple.git
cd HMTntuple/CSCShowerAnalyzer
scram b -j 8
For local testing, do:
cmsRun runCSCShowerAnalyzer_cfg.py inputFiles=file:~/eos/HLT/Commissioning2022/877caf18-4328-415e-bb74-b40adf51f9b4.root runNumber=351407 maxEvents=10
On lxplus, one can submit condor jobs with:
./cmsCondorData.py runCSCShowerAnalyzer_cfg.py $cmssw /eos/cms/store/user/kakwok/HLT/Commissioning2023/run370926 -n 2 -q workday -p /afs/cern.ch/user/k/kakwok/work/private/proxy -i list_370926
./sub_total.jobb
More example commands in ./test/run.sh
To submit jobs on CRAB:
Load crab3 library
source /cvmfs/cms.cern.ch/crab3/crab.sh
The following command is to check crab configuration before submitting. If no issue has been observed, please remove --dryrun
option at the end.
python3 crab_multi_dataset.py -p [pset.py] -o [eos/path/to/save/outputs] -t [tag:str] -i [input text file] --send-external -s [job splitting] -n [unit per job] --work-area [crab directory] --dryrun
The user can also set the run number by -r [run number]
option. Also, if the user is giving private input files list, --private
option can handle the situation. The format of input files is shown in the example directory under test.
Caveat When the user chooses to use --private
option, the job splitting must be set to FileBased!! Also, please check the crab whitelist where to run jobs.
To check the CRAB job status:
python3 crab_multi_dataset.py --work-area [crab directory] --status
The failed job will be automatically resubmitted. If the user doesn't want auto-resubmission, please add --no-resubmit
in the command line.