/AEACuS

Consumer-Level Tools for General Collider Data Analysis, Display, and Optimization

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

AEACuS and RHADAManTHUS

Consumer-Level Tools for General Collider Data Analysis, Display, and Optimization

AEACuS (Algorithmic Event Arbiter and Cut Selector) computes event statistics and performs event selection cuts on event files in the .lhco format.

RHADAManTHUS (Recursively Heuristic Analysis, Display, and Manipulation: The Histogram Utility Suite) does automated plotting of statistics computed by AEACuS.

MInOS (Machine Intelligent Optimization of Statistics) performs automated signal and background discrimination using a Boosted Decision Tree (XGBoost backend).

Enter EXAMPLE directories for five quick-start tutorials, including sample card files for all three programs.

Please contact the author directly ( Joel Walker jwalker@shsu.edu ) with any questions, comments, requests, or invitations to present a talk.

Both programs are included in this directory, and will run without installation or use of any external libraries on any Linux based system (including MacOS) with Perl 5.8+. Windows users may also have success, but this is not officially supported.

A manual for an older version of AEACuS is available here: http://arxiv.org/abs/1207.3383 Note that several newer features are not described, and that certain aspects of the interface have changed. Most critically, program invocation is now: ./aeacus.pl card_file lhco_file cross_section_pb

Slides from a talk on MInOS, AEACuS, and RHADAManTHUS at PHENO 2021 workshop are included with the package distribution Full documentation will be forthcoming on the arXiv at a later date.