A very important part of research in Mathematical Optimization field is to benchmark optimization packages because it is one of the ways to compare solvers. During benchmarking, one usually obtains a large amount of information, like CPU time, number of functions evaluations, number of iterations and much more. This information, if presented as tables, can be difficult to be analyzed and compared, due, for instance, to large amount of data. Therefore, tools to better process and understand optimization benchmark data started to be developed and tested. One of the most widespread tools is the Performance Profile graphics proposed by Dolan and Moré. In this context, this paper describes perprof-py, a free/open source software that makes Performance Profile graphics in a user-friendly manner. This software produces graphics in PDF using LaTeX with PGF/TikZ and PGFPLOTS packages, in PNG using matplotlib, and in HTML using Bokeh. Perprof-py can also be easily extended to be used with other plot libraries, it is implemented in Python 3 with support for internationalization, and is under the General Public License Version 3 (GPLv3).
This is a metapaper for a software we developed while we were working at the Optimization and Operational Research Lab, part of the Department of Applied Mathematic/University of Campinas, Brazil.
View Perprof-py: a Python package for performance profile of mathematical optimization software