The global_optimization package is a Python library designed for restoration of global optimization algorithms.
Currently, the io between ase.Atoms and mealpy project (a library which restores comprehensive global optimization algorithms) has been preliminarily added into this project for further research.
ga: src/ga/ga.py
ssw: src/ssw/main.py
asop: src/asop/asop.py
gwo: src/gwo/GWO.py
mealpy: src/mealpy_io
Here we bind calculator_method:LASP to SSW while calculator_method: MACE to GA. (Currently, our main choice in asop is calculator_method: LASP to SSW, because SSW is a much more efficient global optimization algorithm than GA).
Here, the I/O between ase.Atoms and mealpy has been already written and debugged, my further idea on this may reopen a novel repo for mealpy and that may be called MARGO (Mealpy-based Atomic stRuctural Global Optimization). My further research will be focused on the comparison between selected global optimization algorithms in mealpy and other currently popular methods (including USPEX, Calypso and SSW maybe?). After all these comparison on efficiency and performance, I will valid the perforcement on realistic system (including Surfaces: metal-oxides (metal-inorganic binary systems), metal-metal binary systems, and further Cluster: nanoparticles system).
Here the asop method is just a copyright version by Filend, you may find the official edition from ZhiPan Liu Group.