You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When playing around with the optimizer I stumbled on similar issues as in #1463 .
On Ctrl-C responding with unknown and the best known intermediate solution unfortunately does not yield any value.
Why can't it be sat instead?
Although not optimal, the model clearly is supposed to satisfy the constraints.
When trying to write intermediate solutions or dumping intermediate models nothing happens.
Should be reproducible with this: foo.txt
Version: 4.13.0
The text was updated successfully, but these errors were encountered:
The option :opt.solution_prefix had been bit-rotting. It wasn't enabled.
I have re-enabled it and put it in a place where the models get dumped to files.
There are some other ways to retrieve models:
:opt.dump_models = true - prints the models to the console
Using the programmatic API you can attach a callback when the solver has a model.
I would prefer to return unknown even if there is a model in this case to distinguish optimality from feasibility in the optimization scenario. The optimal remains "unknown".
Hello.
When playing around with the optimizer I stumbled on similar issues as in #1463 .
On
Ctrl-C
responding withunknown
and the best known intermediate solution unfortunately does not yield any value.Why can't it be
sat
instead?Although not optimal, the model clearly is supposed to satisfy the constraints.
When trying to write intermediate solutions or dumping intermediate models nothing happens.
Should be reproducible with this: foo.txt
Version: 4.13.0
The text was updated successfully, but these errors were encountered: