RDataFrame VLQ pair -> single lepton analysis
Replication of step1 for LJMet-Slimmer-1lepDnn analysis with RDataFrame structure for increased readability, and decreased run-time + 'sit-and-wait' time. callRDF.C is used as the command line script that will then call runRDF.C, which houses the main step1RDF_forLJMet.cpp and step1RDF_forLJMet.h files, as well as various .cc files containing functions used within the .cpp. They were moved into .cc scripts to maintain some of the integrity of .cpp and not suffocate the analysis code in functions. The input root files then go through a series of definitions and cuts, with an option to save a snapshot at both preselection and post step1 analysis. This results in a .root file containing the defined branches for plotting and/or furthur analysis.
cmsrel CMSSW_11_0_0
cd CMSSW_11_0_0/src/
cmsenv
scramv1 b
git clone https://github.com/jmhogan/vlq-1lepDnn-RDF
cd vlq-1lepDnn-RDF/
- Change into src area (should be one directory above vlq-1lepDnn-RDF)
cd ../
cp -r ~bluetke/nobackup/YOURWORKINGAREA/CMSSW_11_0_0/src/lwtnn/
scramv1 b
- Find NanoAOD file to use If you intend to save a snapshot, make sure that the final .Snapshot at the very end of the script is uncommented. There's an optional place to start the analysis from a preselection .root file, which cuts down on time significatly. Thus, if you already have preselection root files, I reccomend utilizing that area. If not, the analysis takes anywhere from 30 minutes to 3 hours depending on the size of your input. To run on the command line, use the following:
root -l callRDF.C\(\"Muon(OR)Electron\",\"testNumber\"\)
- Add additional files for condor job submission