Minimum Runtime before considering Gap%
回答済みI have an issue:
I run both slow (hours) and very fast optimization (seconds) and in-between. All run the same model just with different (sizes of) input data.
When i run a slow optimization, a Gap of 5% is completely fine (as more is not really feasible).
When i run a fast optimization I would however prefer Gurobi to try for at least 10s (or until finding the optimal solution) to get to a lower Gap than my specified 5%. Currently it might randomly find a result with a 2% Gap within 1s and then immediately stop.
I don't want to change my Gap goal (to say 0.1%) as I don't always know how fast the optimization will be as it is hard to tell from the data ahead of time.
As far as I can tell there is no way to achieve this currently.
I use pyomo by the way.
Thanks!
-
Did you check whether the ideas in How do I set multiple termination criteria for a model? work for you?
0 -
Interesting, i think i can use that!
Thank you so much!
0
サインインしてコメントを残してください。
コメント
2件のコメント