This is one of my first 'hello world' Haskell projects, a toy solver for the Travelling Salesman Problem.
It uses simulated annealing for optimization and the only metric it optimizes on is Euclidean distance.
As of now there is only one move, a stochastic two-opt move.
Since I wrote this while still learning Haskell, please lower your expectations about code quality, though I don't think it should be too bad.
Algorithmically speaking this is not the best possible algorithm for solving the TSP and there is a lot that can be improved. This is merely a hello world problem for me to learn Haskell.
Credits: Some of the implementation ideas are inspired from this fun tutorial as well as this related talk on Youtube.