Our framework offers state of the art single- and multi-objective algorithms and many more features related to multi-objective optimization such as visualization and decision making. pymoo is available on PyPi and can be installed by:

pip install -U pymoo

Please note, that some modules can be compiled to speed up computations (optional). By using the command above, an attempt is made to compile the modules, however, if unsuccessful the plain python version is installed. More information are available in our Installation Guide.

To get familiar with our framework we recommended having a look at our getting started guide:

Features¶

Furthermore, our framework offers a variety of different features which cover various facets of multi-objective optimization:

Algorithms

Single-objective: Genetic Algorithm, Differential Evolution, Nelder Mead, BRKGA

, CMA-ES

Multi-objective: NSGA-II, R-NSGA-II
Many-objective: NSGA-III, R-NSGA-III, U-NSGA-III, MOEA/D

Test Problems

Single-objective, Multi-objective, Many-objective
Problem Definition, Gradients

Operators

Sampling: Random, LHS
Selection: Random, Binary Tournament
Crossover: SBX, UX, HUX, DE Point, Exponential
Mutation: Polynomial, Bitflip

Visualization

Scatter Plot (2D/3D/ND), Parallel Coordinate Plot (PCP) , Radviz, Star Coordinates, Heatmap, Petal Diagram, Spider Web / Radar, Video

Decision Making

Compromise Programming, Pseudo Weights, High Trade-off Points

Performance Indicator

GD, GD+, IGD, IGD+, Hypervolume, KKTPM

Decomposition

Weighted-Sum, ASF, AASF, Tchebicheff, PBI

Miscellaneous

Termination Criterion, Reference Directions, Constraint Handling, Results

, Callback

, Display

Tutorials

Other Variable Types: Binary, Discrete, Mixed, Custom
More: Parallelization of Function Evaluations, Repair
Examples: Subset Selection

Reference¶

We are currently working on a journal publication. Meanwhile, if you have used our framework for research purposes, you can cite us with:

@misc{pymoo,
    title={pymoo: Multi-objective Optimization in Python},
    author={Julian Blank and Kalyanmoy Deb},
    year={2020},
    eprint={2002.04504},
    archivePrefix={arXiv},
    primaryClass={cs.NE}
}

News¶

April 3, 2020: We are glad to announce pymoo 0.4.0 has been released. We have added two new algorithms (BRKGA, CMAES) and added the test problem suite WFG. Additionally, we have improved the Display Module to create custom printouts and developed a new termination criterion for multi-objective algorithms. (Release Notes)

October 21, 2019: We have released pymoo 0.3.2. We have added one new single-objective optimization algorithm, a performance metric called KKTPM and some new tutorials. (Release Notes)

August 16, 2019: We are glad to announce pymoo 0.3.1 has been released. Many new features were added, for instance visualization and decision making. (Release Notes)

More News

About¶

This framework is developed and maintained by Julian Blank who is affiliated to the Computational Optimization and Innovation Laboratory (COIN) supervised by Kalyanmoy Deb at the Michigan State University in East Lansing, Michigan, USA.

We have developed the framework for research purposes and hope to contribute to the research area by delivering tools for solving and analyzing multi-objective problems. Each algorithm is developed as close as possible to the proposed version to the best of our knowledge. NSGA-II and NSGA-III have been developed collaboratively with one of the authors and, therefore, we recommend using them for official benchmarks.

If you intend to use our framework for any profit-making purposes, please contact us. Also, be aware that even state-of-the-art algorithms are just the starting point for many optimization problems. The full potential of genetic algorithms requires customization and the incorporation of domain knowledge. We have experience for more than 20 years in the optimization field and are eager to tackle challenging problems. Let us know if you are interested in working with experienced collaborators in optimization. Please keep in mind that only through such projects we are able to keep developing and improving our framework and make sure it meets the current needs of the industry.

Moreover, any kind of contribution is more than welcome:

(i) Give us a on GitHub. This makes not only our framework but in general multi-objective optimization more popular by being listed with a higher rank regarding specific keywords.

(ii) In order to offer more and more new algorithms and features, we are more than happy if somebody wants to contribute by developing code. You can see it as a win-win situation, because your development will be linked to your publication(s) which can significantly increase the awareness of your work. Please note that, we aim to keep a high-level of code quality and some refactoring might be suggested. We have prepared a list of suggested contributions.

(iii) You like our framework and you would like to use it for profit-making purposes? We are always searching for industrial collaborations because they help to direct research to meet the needs the of the industry. In our laboratory solving practical problems has a high priority for every student and can help you to benefit from our research experience we have gained over the last years.

If you find a bug or you have any kind of concern regarding correctness please use our Issue Tracker Nobody is perfect and only if we are aware of issues we can start to investigate them.

Content¶

Indices and tables¶

Contact¶

Julian Blank  (blankjul [at] egr.msu.edu)
Michigan State University
Computational Optimization and Innovation Laboratory (COIN)
East Lansing, MI 48824, USA