Greedy circuit cutting and a minimal scheduler based on a Mixed Integer Linear Programming problem.
The project is managed with pdm.
Simply use pdm install to install all necessary dependencies.
We recommend using a virtual environment.
The data module holds the experimental data and setup.
milp.py is a sample using synthetic data.
The benchmark results can be reproduced with python run_experiments.py
MILQ is developed test driven, to see its functionality refer to the test directory.
We use pytest as a testing environment.
MILQ implements the methods described in the following paper.
[1] P. Seitz and M. Geiger and C. B. Mendl. Multithreaded parallelism for heterogeneous clusters of QPUs.