Interesting to see the video, but I suggest that if you're going to demo the TSP, use a graph with >30 nodes for the final demonstration, because there's a O(2^N) algorithm that requires O(2^N) extra space which provides the exact solution, so the various approximate solutions are mostly uninteresting from a performance/practicality point of view. The algorithm is also pretty easy to parallelize, and the amount of extra space can be significantly reduced too. TSP is a pretty interesting problem though so I'm curious to see what comes out.

Yeah the number of nodes is not final yet. I was going to throw a few approximation models at it and demonstrate a few of them in a timely manner. This one algorithm you mention, I want to try this one because that sounds promising.

Ah, yes, referring to the DP solution 🙂 sorry I should have been more precise.