Xpress.jl

Build Status	Social

Xpress.jl is a wrapper for the FICO Xpress Solver.

It has two components:

• a thin wrapper around the complete C API
• an interface to MathOptInterface

The C API can be accessed via Xpress.Lib.XPRSxx functions, where the names and arguments are identical to the C API. See the Xpress documentation for details.

The Xpress wrapper for Julia is community driven and not officially supported by FICO Xpress. If you are a commercial customer interested in official support for Julia from FICO Xpress, let them know!

Install:

Here is the procedure to setup this package:

1. Obtain a license of Xpress and install Xpress solver, following the instructions on FICO's website.

2. Install this package using Pkg.add("Xpress").

3. Make sure the XPRESSDIR environmental variable is set to the path of the Xpress directory. This is part of a standard installation. The Xpress library will be searched for in XPRESSDIR/lib on unix platforms and XPRESSDIR/bin on Windows.

4. Now, you can start using it.

You should use the xpress version matching to your julia installation and vice-versa.

By default, building Xpress.jl will fail if the Xpress library is not found. This may not be desirable in certain cases, for example when part of a package's test suite uses Xpress as an optional test dependency, but Xpress cannot be installed on a CI server running the test suite. To support this use case, the XPRESS_JL_SKIP_LIB_CHECK environment variable may be set (to any value) to make Xpress.jl installable (but not usable).

Use Other Packages

We highly recommend that you use the Xpress.jl package with higher level packages such as JuMP.jl or MathOptInterface.jl.

This can be done using the Xpress.Optimizer object. Here is how to create a JuMP model that uses Xpress as the solver. Parameters are passed as keyword arguments:

using JuMP, Xpress

model = Model(()->Xpress.Optimizer(DEFAULTALG=2, PRESOLVE=0, logfile = "output.log"))

In order to initialize an optimizer without console printing run Xpress.Optimizer(OUTPUTLOG = 0). Setting OUTPUTLOG to zero will also disable printing to the log file in all systems.

For other parameters use Xpress Optimizer manual or type julia -e "using Xpress; println(keys(Xpress.XPRS_ATTRIBUTES))".

If logfile is set to "", log file is disabled and output is printed to the console (there might be issues with console output on windows (it is manually implemented with callbacks)). If logfile is set to a filepath, output is printed to the file. By default, logfile is set to "" (console).

Parameters in a JuMP model can be directly modified:

julia> using Xpress, JuMP;

julia> model = Model(()->Xpress.Optimizer());

julia> get_optimizer_attribute(model, "logfile")

julia> set_optimizer_attribute(model, "logfile", "output.log")

julia> get_optimizer_attribute(model, "logfile")
"output.log"

If you've already created an instance of an MOI Optimizer, you can use MOI.RawParameter to get and set the location of the current logfile.

julia> using Xpress, MathOptInterface; const MOI = MathOptInterface;

julia> OPTIMIZER = Xpress.Optimizer();

julia> MOI.get(OPTIMIZER, MOI.RawParameter("logfile"))
""

julia> MOI.set(OPTIMIZER, MOI.RawParameter("logfile"), "output.log")

julia> MOI.get(OPTIMIZER, MOI.RawParameter("logfile"))
"output.log"

Callbacks

Solver specific and solver independent callbacks are working in MathOptInterface and, consequently, in JuMP. However, the current implementation should be considered experimental.

Environement variables

• XPRESS_JL_SKIP_LIB_CHECK - Used to skip build lib check as previsouly described.

• XPRESS_JL_NO_INFO - Disable license info log.

• XPRESS_JL_NO_DEPS_ERROR - Disable error when do deps.jl file is found.

• XPRESS_JL_NO_AUTO_INIT - Disable automatic run of Xpress.initialize(). Specially useful for explicitly loading the dynamic library.

Julia version warning

The Julia versions 1.1.x do not work properly with MOI dues to Julia bugs. Hence, these versions are not supported.

Benchmarking

In the development of Xpress.jl it is useful to benchmark the MOI wrapper code performance. To perform benchmark we recommend you compare the performance of the master branch aggaints your implementation. Here we leave an example on how to perform the benchmarks the correct way.

1. before starting your implementation run a baseline benchmark aggainst the branch master.

git checkout master
julia --project=benchmark benchmark/benchmark.jl --new bench

2. While testing your implementation benchmark your approach against the baseline benchmark.

git checkout approach_1
julia --project=benchmark benchmark/benchmark.jl --compare bench

When you are ready to make a PR please report the report.txt file content in the PR.

Reference:

FICO optimizer manual