pySLEQP A Sequential Linear Quadratic Programming Method Implemented in Python. We present a prototype implementation of a Sequential Linear Equality-Constrained Qudratic Programming (SLEQP) method for solving the nonlinear programming problem. Similar to SQP active set methods, SLEQP methods are iterative Newton-type methods. In every iteration, a trust region constrained linear programming problem is solved to estimate the active set. Subsequently, a trust region equality constrained quadratic programming problem is solved to obtain a step that promotes locally superlinear convergence. This class of methods has several appealing properties for future research in large-scale nonlinear programming. Implementations of SLEQP methods accessible for research however are scarcely found. To this end, we present pySLEQP , an implementation of an SLEQP method in Python. The performance and robustness of the method and our implementation are assessed using the CUTEst and CUTEr benchmark collections of nonlinear programming problems. pySLEQP is found to show robust behavior and reasonable performance
Keywords for this software
References in zbMATH (referenced in 2 articles )
Showing results 1 to 2 of 2.
- Kirches, Christian; Larson, Jeffrey; Leyffer, Sven; Manns, Paul: Sequential linearization method for bound-constrained mathematical programs with complementarity constraints (2022)
- Lenders, Felix; Kirches, C.; Potschka, A.: \texttttrlib: a vector-free implementation of the GLTR method for iterative solution of the trust region problem (2018)