/desdeo-emo

This repository contains Evolutionary Algorithms that can be used for multi-objective optimization. Interactive optimization is supported. Methods such as RVEA and NSGA-III can be found here.

Primary LanguagePythonMIT LicenseMIT

desdeo-emo

Binder

The evolutionary algorithms package within the DESDEO framework.

Code for the SoftwareX paper can be found in this notebook.

Currently supported:

  • Multi-objective optimization with visualization and interaction support.
  • Preference is accepted as a reference point.
  • Surrogate modelling (neural networks and genetic trees) evolved via EAs.
  • Surrogate assisted optimization
  • Constraint handling using RVEA
  • IOPIS optimization using RVEA and NSGA-III

Currently NOT supported:

  • Binary and integer variables.

To test the code, open the binder link and read example.ipynb.

Read the documentation here

Requirements

Installation process for normal users

  • Create a new virtual enviroment for the project
  • Run: pip install desdeo_emo

Installation process for developers

  • Download and extract the code or git clone
  • Create a new virtual environment for the project
  • Run poetry install inside the virtual environment shell.

Citation

If you decide to use DESDEO is any of your works or research, we would appreciate you citing the appropiate paper published in IEEE Access (open access).