vslch / degree-saturation-algorithm

Degree saturation algorithm for graph coloring problem

degree-saturation-algorithm

Implementation of degree saturation algorithm for graph coloring problem. Two modes are available - greedy and branch-and-bound. Algorithm involves sets intersection operations done via bitwise-logic with custom class TenseBinArray, see tbarray.py.

Usage

color_graph function in dsatur.py

Parameters

• edges - list of tuples (int, int); represent graph's edges;

• blocksize1 - (optional) int, default is 100; parameter for storing binary data associated with vertices, for more details see implementation;

• blocksize2 - (optional) int, default is 100; parameter for storing binary data associated with colors, for more details see implementation;

• mode - (optional) string, available modes are "greedy" and "bnb", default is "greedy"; defines version of algorithm to run

• timeout - (optional) int or None, default is None; defines limit for algorithm's work time in seconds, when exceeded best solution found so far will be returned or None;

• improve - (optional) int or None, default is None; optional parameter for bnb mode, if passed to the function then algorithm will stop working when solution with less or equal colors than G - improve is found where G stands for solution found by greedy approach; if solution with this property is not found, best found solution will be returned.

Returns

• tuple (num, color) where num is number of colors used and color is an array of ints so that color[i] is the color assigned to i'th vertex

Example

>>> from dsatur import color_graph
>>> color_graph([(4,3),(0,1),(1,2),(1,3)])
(2, [1, 0, 1, 1, 0])