Travelling salesman variation in python

[Paul Rubin]

Erlend Andreas Garberg <erlendga at tva.ifi.uio.no> writes:
> An approximate solution would be good enough, do you have any 
> suggestions about how do do that?

Is it just an arbitrary graph?  Most TSP approximation algorithms
depend e.g. on the triangle inequality, i.e. d(A,C) <= d(A,B) + d(B,C).

Papadimitriou and Stieglitz, "Combinatorial Optimization" was a
standard textbook on this stuff, though is pretty old by now.