When using custom branching or node selection rules in SCIPopt, is it possible to force an immediate exit of the branch-and-bound search once a certain known solution is found? I want to say that a specific node is the solution that I want to take, and the B&B procedure should then exit immediately.
I looked at the callback return options for custom branching (https://www.scipopt.org/doc/html/BRANCH.php) and node selection. I don't see an obvious way to return an "exit now". Alternative ways I'm considering: globally fix all the variables or update the global lower bound to match my chosen solution.
Do you want to declare that solution to be optimal somehow? (or do you even know that it is?)
If that is not so important then you could simply callinterruptSolve. Other options would be to set, e.g., the gaplimit to what your gap currently is after finding that solution (using setRealParam)
Edit: So unless you can somehow prove to SCIP that your solution is indeed optimal all your possible options to do this will be somewhat hacky. At that point you might as well force the global lowerbound to match your found solution. Since I don't know how you know that your found solution is optimal I can't really help much more.
Related
Overall, between these two EMR-related commands, another way of looking at this topic/question (to me) is transient vs. long-running.
Based on this research, I'm unsure if create_cluster can do anything that run_job_flow can do (but not the other way around). It seems the main thing to configure is just the KeepJobFlowAliveWhenNoSteps property.
Is what I've stated above valid and complete?
I’m using Optaplanner to make a schedule and it works quite good.
After reading the documentation I have realised that I should use at least 1 (or more) shadow variables since my drool-file is calling methods that does a lot of calculations based on the value of the planningVariable.
I spent a couple of hours rewriting my code to have a shadow variable, but then I notice that the initial solution was really bad (compared to not having shadow variables) and I had to wait severals of minutes just to get an OK result. Is this normal? It did not look like the initial solution used the shadow variable at all.
The question is very generic, and so my answer will be, too.
Sometimes you can simplify the problem by introducing shadow variables or other forms of caching. If you find the right balance, you can indeed speed up the Drools calculation and - as a result - get to the same solution in a shorter amount of time. And therefore, reach better solutions in the same amount of time.
That said, introducing shadow variables shouldn't really change your scores - only how quickly they're calculated. If you're seeing different scores for the same #PlanningSolution, you have in fact changed your problem and the relative performance is no longer comparable.
Also, you may want to check out environment modes to make sure you haven't inadvertently introduced score corruptions into your problem.
I'm building a student schedule generator and I need a way of producing more than one solution. Is there some way to save off feasible scores or scores of Xhard/Ysoft?
I need to be able to output more than one potential schedule, that way the student will have a choice for one schedule over the other if for whatever reason they don't want the "best" schedule (maybe they don't like one of the professors, maybe they don't want an 8am class, whatever)
My original idea was to save off all feasible solutions using the bestSolutionChanged event listener. The problem with this, is that once it finds a 0hard/0soft score, it ignores all scores after that, including scores that are equal.
Ideally I'd like to save off all scores of 0hard/-3soft or better, but just being able to save any feasible scores or force optaplanner to look for a new best score would be useful as well.
This is not a solution, but an analysis of the problem:
Hacking the BestSolutionRecaller is obviously not just a big pain, it's also behaviour we don't want to encourage as it makes upgrading to newer version an even bigger pain. So don't expect us to solve this by adding an easy way to configure that in the solver config any time soon. That being said, a solution for this common problem is clearly needed.
When a new best solution is found, it is planning cloned (see docs for definition) from the working solution (the internal solution in OptaPlanner). This allow us to remember that new best solution as the working solution solution changes. That also means the BestSolutionChangedEvents gets a plannng clone and can safely ship it to another thread, for example to marshal it to a client (presuming any ProblemFactChanges you create do copies instead of alterations), without being corrupted by the solver thread that modifies the working solution.
New best solution imply that workingScore > bestScore. The moment it instead does workingScore >= bestScore, we need far more planning clones (which are a bit CPU expensive), but we could then just send out BestSolutionChangedEvents for that too, if and only if a flag is enabled of course, because most users (unlike yourself) don't want this behaviour.
One proposal is to create a separate BestSolutionChangedOrSameEvent, next to the BestSolutionChangedEvent. This might not be ideal, because we need to be able to detect whether or not someone needs those extra planning clones.
Another proposal is to just have a flag in the <solver> config that switches from > to >= behavior for BestSolutionChangedEvent.
Please create a jira (see "get help" on webpage) and link it it here, or create a support ticket (also see "get help" on webpage).
In the change and swap move selector, I would like to only consider moves that involve entities in conflict as they are more likely to improve the heuristic score.
How should this be done? What classes and interfaces do I have to reuse/extend? I looked at ScoreDirector and PhaseLifecycleListener.
A MoveFilter might do that (if it's not in phase or solver cached as it changes ever step). See the course scheduling example and docs for how to use a filter.
I wouldn't recommend it though, as you still want to move non-conflicting entities at times. You might just want to focus more on those conflicting lectures. So I would keep a vanilla move selector in the mix.
The move filter isn't perfect either - the Guided Local Search feature (not yet available) is a better way to deal with this.
However, given the other question about the model and similar cases I 've seen, I 'd say moves are not your problem. A better model will make all these kinds of move tweaking obsolete.
I am using HP Exstream (formerly Dialogue from Exstream Software) version 5.0.x. It has a feature to define and save boolean expressions as "Rules".
It has been about 6 years since I used this, but does anybody know if you can define a rule in terms of another rule? There is a "VB-like" language in a popup window, so you are not forced to use the and/or, variable-relational expression form, but I don't have documentation handy. :-(
I would like to define a rule, "NotFoo", in terms of "Foo", instead of repeating the inverse of the whole thing. (Yes, that would be retarded, but that's probably what I will be forced to do, as in other examples of what I am maintaining.) Actually, nested rules would have many uses, if I can figure out how to do it.
I later found that what one needs to do in this case is create user defined "functions", which can reference each other (so long as you avoid indirect recursion). Then, use the functions to define the "rules" (and, don't even bother with "library" rules instead of "inline" rules, most of the time).
I'm late to the question but since you had to answer yourself there is a better way to handle it.
The issue with using functions and testing the result is that there's a good chance that you're going to be adding unnecessary processing because the engine will run through the function every time it's called. Not a big issue with a simple function but it can easily become a problem if the function is complex, especially if it's called in several places.
Depending on the timing of the function (you didn't say whether it was a run level, customer level, or specific to particular documents), it's often better to have the function set a User Boolean variable to store the result then in your library rules you can just check the value of the variable without having to run through the function every time.