This project aim to resolve Graph coloring problem using genetic's algorithm. This assignment is a project for Artificial Intelligence class.
Graph coloring consist to assign colors to each vertex , following certain constraints. The minimum colors number adopted to color each node of graph is called chromatic number.
In order to run the project you need to clone it :
git clone https://github.com/borgesis95/leaf-classifier.git
This project was developed using Python3.7 and need some external libraries. You can install them running:
pip3 install -r requirements.txt
After that you can choose from config file (under src folder) several options like which istance (or instances) you want to run, mutation type, crossover type ... etc. In the main folder is also avaiabile a paper which describe more accurately the operator that were developed and why. Right now is available just in Italian.
You can see what happen on each iteration running :
tensorboard --logdir .\logs
Graphic above describe how two different type of configuration works with Dimacs instances le450_15d.col . A the end of iteration, networkx library draw a graph with found solution.