See the wiki for the quick start guide.
HyperMapper is a multi-objective black-box optimization tool based on Bayesian Optimization.
HyperMapper was succesfully applied to real-world problems involving design search spaces with trillions of possible design choices. In particular it was applied to:
- Computer vision and robotics,
- Programming language compilers and hardware design,
- Database management systems (DBMS) parameters configuration.
To learn about the core principles of HyperMapper refer to the papers section at the bottom.
For any questions please contact Luigi Nardi: luigi.nardi at cs.lth.se.
Join the channel for a quicker communication with the dev team:
hypermapper.slack.comHyperMapper is distributed under the MIT license. More information on the license can be found here.
Luigi Nardi, Assistant Professor (Lund University), Research Scientist (Stanford University)
Artur Souza, Ph.D. Student (Federal University of Minas Gerais)
Bruno Bodin, Assistant Professor (National University of Singapore)
Samuel Lundberg (Lund University)
Alfonso White (Imperial College London)
Adel Ejjeh, Ph.D. Student (University of Illinois at Urbana-Champaign)
If you use HyperMapper in scientific publications, we would appreciate citations to the following paper:
Nardi, Luigi, David Koeplinger, and Kunle Olukotun. "Practical Design Space Exploration", IEEE MASCOTS, 2019.
For the list of all publications (including bibtex) related to HyperMapper and its applications, see our Publications page.