rust-gurobi

An unofficial Rust API for Gurobi optimizer.

• Documentation - master / 0.3 (latest)

Notices

• This wrapper library is not officially supported by Gurobi.
• Too many works have not finished yet.

License

Copyright (c) 2016, Yusuke Sasaki

This software is released under the MIT license, see LICENSE.

lib.rs:

This crate provides primitive Rust APIs for Gurobi Optimizer.

It supports some types of mathematical programming problems (e.g. Linear programming; LP, Mixed Integer Linear Programming; MILP, and so on).

Notices

• Before using this crate, you should install Gurobi and obtain a license. The instruction can be found here.

• Make sure that the environment variable GUROBI_HOME is set to the installation path of Gurobi (like C:\gurobi652\win64, /opt/gurobi652/linux64).

• On Windows, the toolchain should be MSVC ABI (it also requires Visual Studio or Visual C++ Build Tools). If you want to use GNU ABI with MinGW-w64/MSYS2 toolchain, you should create the import library for Gurobi runtime DLL (e.g. gurobi65.dll) and put it into GUROBI_HOME/lib. Procedure of creating import library is as follows:

  $ pacman -S mingw-w64-x86_64-tools-git
  $ gendef - $(cygpath $GUROBI_HOME)/bin/gurobi65.dll > gurobi65.def
  $ dlltool --dllname gurobi65.dll --def gurobi65.def --output-lib $(cygpath $GUROBI}HOME)/lib/libgurobi65.dll.a
  

Examples

extern crate gurobi;
use gurobi::*;

fn main() {
  let env = Env::new("logfile.log").unwrap();

  // create an empty model which associated with `env`:
  let mut model = env.new_model("model1").unwrap();

  // add decision variables.
  let x1 = model.add_var("x1", Continuous, 0.0, -INFINITY, INFINITY, &[], &[]).unwrap();
  let x2 = model.add_var("x2", Integer, 0.0, -INFINITY, INFINITY, &[], &[]).unwrap();

  // integrate all of the variables into the model.
  model.update().unwrap();

  // add a linear constraint
  model.add_constr("c0", &x1 + 2.0 * &x2, Greater, -14.0).unwrap();
  model.add_constr("c1", -4.0 * &x1 - 1.0 * &x2, Less, -33.0).unwrap();
  model.add_constr("c2", 2.0 * &x1 + &x2, Less, 20.0).unwrap();

  // integrate all of the constraints into the model.
  model.update().unwrap();

  // set the expression of objective function.
  model.set_objective(8.0 * &x1 + &x2, Minimize).unwrap();

  assert_eq!(model.get(attr::IsMIP).unwrap(), 1, "Model is not a MIP.");

  // write constructed model to the file.
  model.write("logfile.lp").unwrap();

  // optimize the model.
  model.optimize().unwrap();
  assert_eq!(model.status().unwrap(), Status::Optimal);

  assert_eq!(model.get(attr::ObjVal).unwrap() , 59.0);

  let val = model.get_values(attr::X, &[x1, x2]).unwrap();
  assert_eq!(val, [6.5, 7.0]);
}