Compute Metaheuristic Engine for Python (cmenpy or /siːmɛnpaɪ/) is an extensible, flexible and vectorized Python framework explicitly architected for designing and prototyping metaheuristic and population-based optimization algorithms.
It provides the clean, abstractions necessary to build custom evolutionary algorithms, swarm intelligence methods, and hybrid heuristics from scratch, granting you absolute architectural control.
- Vectorized Data Structures: Built from the ground up to maximize data locality and mathematical efficiency. Population arrays, agent states, and search spaces leverage a completely flattened, vectorized structure that eliminates heavy object overhead and interfaces natively with NumPy and parallel computing paradigms.
- Extensible Architecture: Designed with clean, highly pedagogical abstractions. Crafting custom transition operators, custom mutation rules, or completely new optimization paradigms requires overriding just a few decoupled, well-defined methods.
- Non-Monolithic Flexibility: We cut out all the bloat. Bypasses secondary overhead—such as visualization tools, complex file exporters, and rigid internal loggers—focusing entirely on the raw mathematical engine and structural flow.
- Research Ready: Protected by the GNU GPLv3 License. You have complete freedom to design, modify, and distribute your custom algorithms. Any derivative work or integration must also remain open and free, ensuring the software permanently evolves as a public good for both academic research and open commercial development.
- Developed by: LTSIM @ 2026