SuperSCS is is a fast and accurate solver for conic optimization problems, that is, problems of the form
minimize c'x
subject to Ax + s = b
s in K
where K is a nonempty, closed, convex cone.
It can be accessed from MATLAB and Python, directly, or via CVX/CVXPy.
Detailed documentation can be found here.
Download the current stable version
Check out the installation instructions
It is straightforward to download and run the SuperSCS docker image.
Simply run:
docker pull kulforbes/superscs
docker run -it kulforbes/superscs
Find out more information here.
SuperSCS can be used in the following ways
- Using the SuperSCS C library
- in MATLAB
- directly
- via CVX
- in Python
- directly
- via CVXPy
Read the documentation for further information.
SuperSCS is based on the SuperMann algorithmic scheme; plese, cite as follows:
A. Themelis and P. Patrinos, "SuperMann: a superlinearly convergent algorithm for finding fixed points of nonexpansive operators," arXiv:1609.06955, 2017.
Cite the software as follows:
@MISC{superscs,
AUTHOR = {Sopasakis, P. and Menounou, K. and Patrinos, P.},
HOWPUBLISHED = {\url{https://kul-optec.github.io/scs/}},
MONTH = {Apr},
TITLE = {{SuperSCS}: A fast and accurate conic optimization solver},
YEAR = {2017},
}