Skip to main content

Phoenix for Event Visualisation

Visualizing an event inside the detector opens a path to a deeper understanding of the physical processes resulting from a collision. One commonly used tool for event visualisation is Phoenix, a framework that allows 3D visualisation of collision events inside the detector.

The event in the image above is an event for gluino production, which in SUSY is the supersymmetric partner of a gluon.

You can display any event of your choice by using the open data available in PHYSLITE format, and the process to do so is explained below:

Transforming PHYSLITE or NTuple into JSON

Phoenix accepts data in JSON or JiveXML formats. We will focus on the JSON format. To create the JSON from a PHYSLITE or NTuple file, we provide a python script:

To use the code, follow the steps below:

  1. Download a PHYSLITE or NTuple file: The datasets can be found in the CERN Open Data portal, and the instructions on how to search for one of our PHYSLITE datasets are in Accessing the Data.

  2. Get code: You can download the script and enviroment file, or clone the whole repository:

    git clone https://github.com/atlas-outreach-data-tools/notebooks-collection-opendata.git
  3. Setup enviroment: to setup the enviroment you can use the enviroment file with Conda:

    conda env create -f path_to_file/environment_phoenix.yml

    Make sure to adjust path_to_file to the path where the code and the enviroment file are. If you don't use conda, these are the main libraries you will need:

    numpy==1.22.4
    uproot==4.3.7
    awkward==1.10.5
    coffea==0.7.21
    zstandard==0.22.0
  4. Run the script: To see the options of the script, you can run

    python path_to_file/get_json_phoenix.yml --help

    And you will see the options for running:

    usage: write_simple_json.py [-h] [-n NEVENTS] [-s NSKIP] [-o OUTPUTFILE]
    [-l EVENTLIST]
    filenames

    Writes events in simple json format for the Phoenix event display

    positional arguments:
    filenames Comma-separated list of files for processing

    optional arguments:
    -h, --help show this help message and exit
    -n NEVENTS, --nEvents NEVENTS
    Number of events to run over (default 10)
    -s NSKIP, --nSkip NSKIP
    Number of events to skip (default 0)
    -o OUTPUTFILE, --outputFile OUTPUTFILE
    Name of output json file (default my_events.json)
    -l EVENTLIST, --eventList EVENTLIST
    File containing run/event number pairs for outputting

    For example, for the gluino display, you can run the following code to generate the JSON file using 20 events:

    python get_json_phoenix.py path_to_file/mc20_13TeV.376518.MadGraphPythia8EvtGen_A14NNPDF23LO_GG_direct_2400_200.deriv.DAOD_PHYSLITE.e6732_s3797_r13167_p6026/DAOD_PHYSLITE.38191276._000001.pool.root.1 -n 20

    After running, you should have my_events.json in the folder where you ran the script, which now can be uploaded to Phoenix.

You can open your own files in Phoenix using the file upload dialog (third, right to left, in the figure below)

You can open your own files using the File upload dialog (third, right to left)
Fig. 1: Phoenix bar. The file upload dialog is the one inside the red square.

After that, you should see your event!

For more information about the use of Phoenix, see the Phoenix User Guide.