This repository is intended to setup a software workflow for using the EIC smear package to generate pythia truth MC and "smear" it using a fast detector simulation.
This has only been tested on RCF, where access to the EIC environment is present. In order to use this, it is expected you have setup this environment via
$ setenv EIC_LEVEL pro
$ source /afs/rhic.bnl.gov/eic/restructured/etc/eic_cshrc.csh
In principle, if you have EICSmear and ROOT installed on your local computer with a similar environment setup to the RCF version, this code would be functional. One would have to change, in the scripts, the directory where their EIC environment setup script exists.
All that is necessary to use this is to run the main runSimWorkflow.py with the arguments as documented in the file. Alternatively, there is a .csh script in condorEventGen that runs the same thing - this may be more intuitive for some users. The output is a truth root file, which contains a tree with truth particle information from PYTHIA, and a smeared root file, which contains the particles after undergoing detector smearing.
The detector smearing is set in smear/ePHENIXDetector.cpp and smear/smearHandBook.cxx.
The output truth and smeared files can be befriended for further analysis within ROOT, as in the directory analysisCode.
An example command to run the python script would be
python runSimWorkflow.py smeared truth 275 18 9 100 /path/to/directory/EICJetSubstructure/ 0
This would run the script and produce a file smeared.root and truth.root which contain the truth and smeared output from EICSmear. The proton and electron energy would be 275 and 18 GeV, respectively, with a minimum Q2 of 9. The job would process 100 events. The path argument allows python to find the necessary files it needs to run, and should just be the global path to where you cloned the repository. The final integer is a number which identifies the job, if running on condor (otherwise it is arbitrary).
The repository has three main analysis steps - event generation, event analysis, and macro. Each step and the relevant location of the code are described in more detail here. An example single DIS event truth and smeared output can be found in MCData/example, which corresponds to the event generation output for 1 event. In principle, one could run the analysis chain on this example output (although you won't get any interesting results!).
The event generation is controlled by the runSimWorkflow.py script, which calls separate python scripts in the directories pythiaGeneration and smear to do the truth event generation and smearing, respectively. For those working on RCF who have access to the condor node farm, condor scripts are available to run the event generation in condorEventGen.
After a truth and smeared file are created, the code to analyze the output events is available in the analysisCode directory. Please see that directory for another README describing its use. A basic example analysis code with Makefile can be found in the example directory, should you want to create your own analysis code.
Macros to draw histograms from the output event analysis trees are stored in analysisCode/macros
Don't be alarmed if you see "Error in TROOT::TVector2" type errors in the smearing. The smearing process sometimes removes a particle completely (e.g. for efficiency loss) and thus root complains when trying to smear a nonexistent particle.
To execute the runSimWorkflow.py, you can run the following:
python runSimWorkflow.py smeared truth 100 10 9 1000 /base/path/to/your/jetSubstructure 0
This corresponds to running the workflow and generating a smeared file called “smeared.root”, a truth file called “truth.root”, proton energy = 100 GeV, electron energy = 10 Gev, minimum Q2=9 GeV, and for 1000 events. The path argument is the base path on RCF to wherever you cloned this jetSubstructure repository - for example, /User/joe/git/jetSubstructure. The final value is an arbitrary number indicating the process ID - this is relevant for running on condor when running multiple batch jobs.
RCF also has a condor batch processing system available. Condor submission jobs are available in the condorEventGen directory. The csh script can be run with similar arguments to the python runSimWorkflow.py script. The condor .job file is used to submit jobs to condor and can be used with the usual condor calls. See also the README in the directory condorEventGen directory.
For additional information regarding EICSmear, see here.