Scripts in Python and R that calculate ranking and the percentage performance indicators of metaheuristic optimizers (the percentage of trials that found the global optimum, the percentage of thresholds achieved by all trials, and the percentage of budget left). Typical rankings used during optimization competitions are sensitive to tiny changes in the error level (epsilon change switches the rank); some do not consider speed, and some treat speed as equally important to finding the global optimum. The detailed justification and description are in the paper. These scripts are configured to work with data from CEC2022. The exemplary data for comparison of the two algorithms
NL-SHADE-LBC parametrized - C++ code. The original code was downloaded from CEC2022 competition organizers. In the modified version, all parameters hidden as numerical constants in the source code were found and named; they can be set from the command line. A more detailed explanation is in the paper.
S-LSHADE-DP parametrized - C++ code. The original code was downloaded from CEC2022 competition organizers. In the modified version, all parameters hidden as numerical constants in the source code were found and named; they can be set from the command line. A more detailed explanation is in the paper.
NL-SHADE-RSP-MID parametrized - C++ code. The algorithm took third place in the CEC2022 competition on single objective bound constrained search. As the name reveals, it is the modification of NL-SHADE-RSP, which
was downloaded from Suganthan's repository. A more detailed explanation of the parametrization is in the paper. The description of the algorithm is in the paper
Corrected EA4eig - Octave/Matlab code. The original code was downloaded from CEC2022 competition organizers. In the original version, the decision path that used jSO and Eigen crossover did not respect the bounds of the search space. A more detailed explanation is in the paper.
The code of RB-IPOP-CMA-ES in R language (my version of IPOP (based on cmaesr package) was cleaned and partially rewritten by a Ph.D. student). A more detailed explanation is in the paper.