A demo on using the D-Wave Ocean SDK to solve the map coloring problem. Namely, given a map, color the regions of the map such that no two regions sharing a border would share the same color.
A graph representation of the provinces of Canada. Provinces connected by edges share a border together. This is a sample output produced by this demo.
Simply run the code with
python map_coloring.pyThe idea is to describe the map coloring problem in terms of constraints. Once this is done, we can use tools from the Ocean SDK to convert these constraints into a binary quadratic model (BQM), a type of equation that can be ingested by the quantum computer. Afterwards, we will hopefully have a solution to our map coloring problem.
Constraints to describe the map coloring problem:
- Each region can only select one color
- No regions sharing a border can share a color
- In the code, we let each of the regions choose one among four colors. Why are we limiting ourselves to only four colors? According to the Four Colour Theorem, we need no more than four colors to color any planar map such that no two adjacent regions share the same color
- To solve the map coloring problem, in this demo, we are forming constraints, converting said constraints into a BQM, and then feeding said BQM to a solver. Please note, however, that the D-Wave Ocean stack also has a function called min_vertex_color_qubo(..) that will directly form the map coloring BQM for you.
Released under the Apache License 2.0. See LICENSE file.