Kona is a library for nonlinear constrained optimization.
Kona was designed primarily for large-scale partial-differential-equation (PDE) governed optimization problems; however it is suitable for any (sufficiently smooth) problem where the objective function and/or constraints require the solution of a computational expensive state equation.
As a consequence of its abstracted vector and matrix implementations, Kona is also useful for developing new optimization algorithms for PDE-governed optimization.
Please refer to the code documentation, API reference and use examples on ReadTheDocs for more details on Kona's parallel-agnostic implementation and linear algebra abstraction.
Kona was originally written by Dr. Jason E. Hicken in C++. This old version is now deprecated, but still available on BitBucket.
The Python re-write of Kona is being developed and maintained by Dr. Hicken and the Optimal Design Lab research group at Rensselaer Polytechnic Institute.