Skip to content

Latest commit

 

History

History
79 lines (59 loc) · 2.15 KB

SOLNP_installation.md

File metadata and controls

79 lines (59 loc) · 2.15 KB
Raw

Requirements

CMake >= 3.5.1

MKL or LAPACK

OSQP

Installation

Currently, we provide C, MATLAB, Python and Julia interface on macOS, Linux and Windows.

In terms of the linear algebra library, we provide support for both Intel MKL and open-source LAPACK, which should be installed in advance. On macOS, there is no need to install any linear algebra library, as we leverage Apple's proprietary Accelerate framework. On Linux and Windows, you can decide whether to use MKL by configuring the parameter LINK_MKL in CMakeLists.txt. If LAPACK or MKL can not be found automatically, you can specify the path in CMakeLists.txt manually. Before installation, you also need to specify the environment variables of OSQP directories:

export OSQP_HOME=your_path_to_osqp

We compile MATLAB interface within MATLAB, you can simply run the following command in MATLAB command line:

cd SOLNP_PLUS_ROOT
cd interface/Matlab
make_solnp

As for C, Python and Julia interface, you need to first compile SOLNP+ source code:

On Linux and macOS :

cd SOLNP_PLUS_ROOT
mkdir build
cd build
cmake ..
cmake --build .

On Windows:

cd SOLNP_PLUS_ROOT
mkdir build
cd build
cmake -G "MinGW Makefiles" ..
cmake --build .

After compiling source code, you can directly use the C interface. We give CMakeLists.txt and a simple example, from which you can compile executable file:

cd SOLNP_PLUS_ROOT
cd interface/C/build
cmake ..
cmake --build .
./solnp_demo

You can also Install and test Python interface:

cd SOLNP_PLUS_ROOT
cd interface/Python
python setup.py install
cd test
python test.py

And Julia interface:

cd SOLNP_PLUS_ROOT
cd interface/Julia
  • Type in julia in command line to start a Julia REPL. Then type in ] to enter Pkg mode and type in dev ./SOLNP/. Type in [Backspace] to leave Pkg mode.
  • To test whether everything is ready, try to run test file in Julia REPL by typing in
julia> include("./SOLNP/test/test.jl")