This repository contains an Answer Set Programming (ASP) implementation using Clingo for scheduling the LZV Cup, an amateur indoor football league in Belgium.
Seems interesting. Quite a bit of thought would have to go into how variables are defined however - such that their domains can be set nicely. Failing to see how this could be bootstrapped onto the current base system, would need to redesign from the top I believe?
Any simple ideas of how to model like this? Would we be trying to just implement the IP constraints from the paper? If so, it seems like using an actual MIPs solver makes more sense? What is your intuition here?
Random aside news:
Interesting I discovered that clingcon cannot enumerate optimal models when asked - instead you need to control it using the python API and print all solns with a certain bound. My example program that showed this:
&dom {0..1} = x.
&dom {0..1} = y.
&distinct { x ; y }.
&minimize { x ; y }.
It should obviously give two minimums: x=1,y=0 and x=0,y=1.
I had a play with clingcon on some dummy examples - https://github.com/potassco/clingcon/tree/master/examples
Seems interesting. Quite a bit of thought would have to go into how variables are defined however - such that their domains can be set nicely. Failing to see how this could be bootstrapped onto the current base system, would need to redesign from the top I believe?
Any simple ideas of how to model like this? Would we be trying to just implement the IP constraints from the paper? If so, it seems like using an actual MIPs solver makes more sense? What is your intuition here?
Random aside news: Interesting I discovered that clingcon cannot enumerate optimal models when asked - instead you need to control it using the python API and print all solns with a certain bound. My example program that showed this:
It should obviously give two minimums: x=1,y=0 and x=0,y=1.
The issue is discussed in https://github.com/potassco/clingcon/issues/78 if you are interested. Took me a little while to find it.
Originally posted by @axelahmer in https://github.com/axelahmer/lzcup-scheduler/issues/2#issuecomment-2158073908