Lorenzett Event Simulator

This is the first event generator for HEP using a generic calorimeter for machine learning studies based on athena framework.

Getting Started

There are two ways you can run Lorenzett: using a Docker container (recommended) or locally on your machine.

Running with Docker

If you take a look at the Lorenzett's DockerHub you'll find two main images:

• gabrielmilan/lorenzett:latest
• gabrielmilan/lorenzett:sdumont

We'll refer to the latest image as base, since sdumont inherits from it. More details on how these images are generated here, on the docker/ directory.

Our base image is meant for users that want to do a custom run on Lorenzett. If you're a first-timer, you'll want the gabrielmilan/lorenzett:sdumont image.

Using the sdumont image

docker run -v <output-path>:/output gabrielmilan/lorenzett:sdumont <args>

For <args> information, see below:

lorenzett_run.sh - Script for running Lorenzett through docker in an one-line script
 
./lorenzett_run.sh [arguments]
 
Arguments:
-h, --help                show this message
-f, --filter              chooses which type of event you're interested
-e, --event               sets the desired number of events
-p, --pileup              sets the pileup average
-o, --output              sets the output filename
--bc_id_start             sets bc_id_start
--bc_id_end               sets bc_id_end
 
All arguments are required

Using the latest image

docker run -it -v <your-mount>:/volume gabrielmilan/lorenzett /bin/bash

Note: You shall mount as many volumes as you wish.

As soon as your bash opens, you need to run

source setup_envs.sh

This will setup everything you need for Lorenzett.

For event generation, this is the command you're looking for (example for Zee):

prun_job.py -c "generator.py --filter Zee -i zee_config.cmnd --outputLevel 6 --seed 0 -evt <n_events> --pileupAvg <average-pileup> --bc_id_start -8 --bc_id_end 7" -o zee.root -mt <n_threads> -n <n_jobs>

Notes

• Files like zee_config.cmnd are located at /code/lorenzett/generator/PythiaGenerator/data/
• Using seed=0 means that you'll use your system clock as seed.
• If you would like to collect minimum bias for an especific calorimete region you just include the eta/phi region here and use the --filter with name MinimumBias.

For reconstruction, after generate the events using the generator.py command, you must pass the output file as input to the reconstruction transformation. To run the reconstruction use this command:

reco_trf.py -i zee.root --outputLevel 6 -nt <n_jobs> -o reco_zee.root

Running locally

Requirements

• Geant4 (opengl or qt4 is required for graphic interface, https://github.com/jodafons/geant4.git);
• ROOT (https://github.com/root-project/root.git);
• Pythia8 (https://github.com/jodafons/pythia.git);
• HEPMC (https://github.com/jodafons/hepmc.git);
• FastJet (https://github.com/jodafons/fastjet.git).

References:

• Athena framework: https://gitlab.cern.ch/atlas/athena