1246. Recent new features of FICO Xpress APIs and Xpress Mosel
Invited abstract in session WA-43: Modelling languages, stream Software for Optimization.
Wednesday, 8:30-10:00Room: Newlyn GR.07
Authors (first author is the speaker)
| 1. | Susanne Heipcke
|
| Xpress Optimization, FICO |
Abstract
This talk gives an overview of recent updates to the FICO Xpress APIs: a completely new set of object-oriented APIs (C++, C#, Java) has been published that make use of modern programming language features. These APIs work directly with Xpress Solver without any need for keeping intermediate copies of problem data.
The Python API for Xpress has been reworked to align it more closely with the other APIs, in particular by introducing linked objects that directly modify the problem, leading to performance improvements for large problems. Besides NumPy, the Python API now also supports SciPy for the efficient creation of large constraint expressions.
We shall also give a short introduction to the newly published Julia API for Xpress that is provided open-source on Github (see https://github.com/fico-xpress/XpressAPI.jl) and through the official Julia registry.
FICO Xpress Mosel has undergone some restructuring of its distribution and build processes, with significant updates to its open source component 'moseltest' (see https://github.com/fico-xpress/mosel). Other new Mosel features include improvements to data handling functionality and support of new Xpress features.
Keywords
- Modeling Systems and Languages
- Software
- Optimization Modeling
Status: accepted
Back to the list of papers