Skip to content

Conversation

odow
Copy link
Member

@odow odow commented Dec 30, 2024

Extracted from #2596

The behaviour should, where possible, not depend on this somewhat arbitrary order of bridges in this function. That is too brittle and the behaviour is undocumented.

const RSOCtoPSD{T,OT<:MOI.ModelLike} =
SingleBridgeOptimizer{RSOCtoPSDBridge{T},OT}

MOI.Bridges.bridging_cost(::Type{<:RSOCtoPSDBridge}) = 1.5
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So there's an argument to be made here. Would it be better to globally penalise PSD sets more than RSOC sets? We're starting to muck with bridge costs, and it becomes a bit subjective.

Copy link
Member

@blegat blegat Jan 2, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's debatable, this one is clearly destructing structure so it make sense to give it a high cost. But this would be inferrable form a higher cost given to PSD variables. You can even give a higher cost than 1.5 like 5 or 10, it should really only be used when there is no other choice ^^

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've changed to 10

@odow odow merged commit 5feb1c0 into master Jan 2, 2025
16 checks passed
@odow odow deleted the od/variable branch January 2, 2025 22:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

2 participants