FAST-HEP is a group of High Energy Physics (HEP) researchers who:

• Contribute to existing software tools and/or develop its own packages to close gaps in the High Energy Physics software ecosystem.
• Develop and showcase an “ideal FAST-HEP analysis” and demonstrate its capabilities and performance.
• Provide expertise on how to incorporate modern analysis techniques into analyses (eg. deep machine learning, vectorised array programming, etc), and provides feedback to the developers of such tools.
• Educate FAST-HEP members and colleagues on what modern analysis tools and techniques are available and encourages knowledge sharing through presentations and workshops.

The current set of interacting packages is described in this figure:

The packages are available on pypi and their code on GitHub. Installation can be achieved through, for example, the fasthep meta-package:

  pip install fasthep[full]

FAST-HEP is currently used by members of: the CMS Collaboration (including analysis and upgrade work), the LZ Dark Matter Experiment, and the Deep Underground Neutrino Experiment (DUNE). Collaborations or individuals that are interested in our tools or need support can reach us at fast-hep@cern.ch. We also have a community on gitter: https://gitter.im/FAST-HEP/community.

The tools are written in python and are powered by external libraries and tools: pandas, numpy, matplotlib, scikit-hep (in particular uproot), Dask, etc.