tpmccauley/workshop2022-lesson-ttbarljetsanalysis-payload

TTBarLJets analysis using Coffea

This example is an adaptation of the example analysis presented at the IRIS-HEP AGC Workwhop 2022

Additional packages

In order to run the simplified analysis with coffea, we use the python tools docker container, which was prepared for the CMS Open Data Workshop 2022.

This container alread had uproot, awkward, numpy and xrootd installed. We need a few extra tools. One can install them, directly into the container with:

pip install vector hist mplhep coffea

Working examples

• To run a very simple version of the essence of this analysis, without systematics, and over one simple dataseta and analysis region in order to produce a single observable plot:

  python coffeaAnalysis_basics.py

  Ordered pieces of it, from top to bottom, can be better run on a jupyter notebook. Or, if preferred, run interactively:

  python -i coffeaAnalysis_basics.py

• To run the full analysis but over a limited ntuples.json file (with just one input file per dataset), do:

  python coffeaAnalysis_ttbarljets.py

  Final images for histograms are not saved well, but the final histograms.root should be sound. This is the file which will be used later with cabinetry.