Pyevolve in action, solving a 100 cities TSP problem
Here is the video of Pyevolve (the development version) optimizing a 100 cities TSP problem. I’ve used the Edge Recombination and the simple Swap Mutation method:
The video is a composition the of image outputs at every 100th generation and only when the score has changed when compared to the previous generation. I’ll release the source together with the new 0.6 release, planned to this month (December) and no later than January.
Best wishes and if I can’t post again until the next year, merry christmas and happy new year !!!
– Christian S. Perone
This is exciting. I’m looking forward to trying out the new distributed functions on a cluster of Amazon EC2 instances. Thanks again.
An optimal TSP solution shouldn’t have crossings. Example 12 of pyevolve also shows a nonoptimal solution.
Hello Wurstmann, I didn’t waited for an optimal solution.