CAVE stands for Configuration Visualization, Assessment and Evaluation. It is a versatile analysis tool for automatic algorithm configurators. It generates comprehensive reports (e.g. that give you insights into the configured algorithm, the used instance set and also the configuration tool itself. The current version works out-of-the-box with SMAC3, but can be easily adapted to other configurators, as CAVE supports CSV files as input.


CAVE can be obtained on our github page (external link) or through pypi (external link). Documentation for CAVE is available here (external link).


Biedenkapp, A. and Marben, J. and Lindauer, M. and Hutter, F. (pdf)(bib)
CAVE: Configuration Assessment, Visualization and Evaluation
In: Proceedings of the International Conference on Learning and Intelligent Optimization (LION’18)
To appear