N-Wouda / Euro-NeurIPS-2022

OptiML's contribution to the EURO meets NeurIPS 2022 vehicle routing competition.
Other
16 stars 2 forks source link

Configuration management #118

Closed N-Wouda closed 2 years ago

N-Wouda commented 2 years ago

We have quite a few parameters in the static solver and the rollout algorithm. Right now these are managed fairly ad-hoc:

We should streamline this a bit more. I'm thinking of a general Config class on the Python side, that stores the relevant static operators, and the static and dynamic parameters. It should support being constructed from code (for tuning), or by reading from a file (e.g., with a classmethod from_file).