MadGraph5_aMC-NLO Python 3 Docker Image

Docker image for Python 3 compliant MadGraph5_aMC@NLO.

Distributed Software

The Docker image contains:

• MadGraph5_aMC@NLO v3.5.1
• Python 3.10 (3.9 CentOS)
• HepMC2 v2.06.11
• LHAPDF v6.5.3
• FastJet v3.3.4
• PYTHIA v8.306
• BOOST v1.76.0

Additionally contains MadGraph5 controlled dependencies for NLO processes:

• CutTools
• IREGI
• Ninja
• Collier

Installation

• Check the list of available tags on Docker Hub to find the tag you want.
• Use docker pull to pull down the image corresponding to the tag. For example:

docker pull scailfin/madgraph5-amc-nlo:mg5_amc3.5.1

Use

MadGraph5_aMC@NLO is in PATH when the container starts

docker run --rm scailfin/madgraph5-amc-nlo:mg5_amc3.5.1 "mg5_aMC --help"
Usage: mg5_aMC [options] [FILE]

Options:
  -h, --help            show this help message and exit
  -l LOGGING, --logging=LOGGING
                        logging level (DEBUG|INFO|WARNING|ERROR|CRITICAL)
                        [INFO]
  -f FILE, --file=FILE  Use script file FILE
  -d MGME_DIR, --mgme_dir=MGME_DIR
                        Use MG_ME directory MGME_DIR
  --web                 force to be in secure mode
  --debug               force to launch debug mode
  -m PLUGIN, --mode=PLUGIN
                        Define some additional command provide by a PLUGIN
  -s, --nocaffeinate    For mac user, forbids to use caffeinate when running
                        with a script

so you should be able to make any directory inside the container a working directory and run mg5_aMC commands from there.

If you run the image as an interactive container with your local path bind mounted to the working directory

docker run --rm -ti -v $PWD:$PWD -w $PWD scailfin/madgraph5-amc-nlo:mg5_amc3.5.1

output from your work in that directory in the interactive session will be preserved when the container exists.

The container can be used a runtime application by passing in a MadGraph program as a .mg5 file to the mg5_aMC CLI API

docker run --rm -v $PWD:$PWD -w $PWD scailfin/madgraph5-amc-nlo:mg5_amc3.5.1 "mg5_aMC file-name.mg5"

For further examples see the tests.

Tests

As an example test you can run the top mass scan example in the tests directory inside the Docker container by running the following from the top level directory of this repository

docker run --rm -v $PWD:$PWD scailfin/madgraph5-amc-nlo:mg5_amc3.5.1 "cd ${PWD}/tests; bash tests.sh"

or run the test runner

bash test_runner.sh