Analyzing the Impact of Undersampling on the Benchmarkingand Configuration of Evolutionary Algorithms - Dataset

This repository contains the raw data and code nessecary to reproduce the results from the paper "Analyzing the Impact of Undersampling on the Benchmarkingand Configuration of Evolutionary Algorithms"

The main file is the python-notebook 'reproducibility.ipynb', which details the full process for reproduction of the results shown in the paper. The two additional .py files are included for computation which takes longer and can be parallelized.

The folder 'irace_conf_static_modcma.zip' contains the verification runs: 200 independent runs of each configuration. Indexes are according to 'Irace_confs_static_modcma_v2.csv'

The folder 'logs_baseline_cs.zip' contains the raw irace files on which the analysis is based. This data is taken from the following repository:
de Nobel, Jacob, Vermetten, Diederick, Wang, Hao, Doerr, Carola, & Bäck, Thomas. (2021). Data and Code from: Tuning as a means of assessing the benefits of new ideas in interplay with existing algorithmic modules (1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.4524959