Michael Feldmeier
Michael Feldmeier
Thanks for reporting this issue. Yes, the .lp filereader could be more efficient, but that would require a complete rewrite taking up amount of time I don't have at the...
No, this is not a priority - thanks for your updates to the GAMS interface!
Thanks for spotting this and providing the QP, I'll investigate!
Sorry for the slow response. I couldn't reproduce the issue. This is likely due to a bug in the .lp writing procedure in HiGHS that was fixed recently- would you...
Thanks, will try to reproduce and fix
The loop arises as the QP solver wrongly thinks that the search direction in the second iteration doesn't have curvature (p'Qp is less than a threshold) and thus oversteps. It...
I'll see what I can do. Detecting overstepping (objective getting worse) is easy, Detecting a loop of length 2 should be doable, but arbitrary length loops without objective change takes...
Tightening the tolerance doesn't appear to break any other QP instances. I'll leave cycle detection for when I'm done with my dissertation
This bug is still around. Stumbled across it today. Our workaround is reapeating the bounds as constraints.
This might be a side effect I didn't anticipate of a recent improvement I made for unconstrained QPs. I'll check