Tools setup for Math 381

We will be using the following tools

• Python
• SCIP (linear programming solver)
• JupyterLab

Method 1: The Quickest Way:

Follow this guide. In particular install

1. Docker for Windows/Mac. For linux read the guide above.
2. Run docker run --rm -p 9001:9001 scipoptsuite/scipoptsuite:7.0.2-teach

Caveats: Docker is a very portable way to get started. However, each time you quit and restart docker, it resets everything. So you might have to download your files and upload. Read usage notes in the guide above.

Method 2: The slower but more permanent way:

• I assume you have python installed. Remember we are using Python 3+.

• Then you need to install conda

• Finally install PySCIPOpt via conda install --channel conda-forge pyscipopt

• Install JupyterLab via pip install jupyterlab

Mathod 3: Google Colab way

• Go to colab.research.google.com and get started.

Google Colab allows you a nice cloud computing (free) setup with python and jupyter. Unfortunately the PySCIPOpt module (for solving Linear Programs) does not come preinstalled.

• In order to install PySCIPOpt run this (Shift+Enter) inside a jupyter notebook whenever you create a new notebook.

!pip install -q condacolab
import condacolab
condacolab.install()
!conda install pyscipopt

• Finally check if everything works by this program

from pyscipopt import Model
model = Model("Example")  # model name is optional
x = model.addVar("x")
y = model.addVar("y", vtype="INTEGER")
model.setObjective(x + y)
model.addCons(2*x - y*y >= 0)
model.optimize()
sol = model.getBestSol()
print("x: {}".format(sol[x]))
print("y: {}".format(sol[y]))

You should get the output

x: 0.0
y: 0.0

Some Remarks

• The programming language recommended is Python, however you are free to use your own preferred language and solver.
• For example you can use lpsolve as well.
• Another nice solver is ortools from Google.