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
Is there a reason to not allow eps_abs positive and eps_rel zero? Many other QP solvers allow you to enforce an absolute tolerance on constraints, so setting eps_rel to zero would help when comparing OSQP to these solvers. Of course, it's possible to set it to an extremely tiny value, but you shouldn't need to do this.
The text was updated successfully, but these errors were encountered:
OSQP currently enforces that eps_rel be positive:
https://github.com/oxfordcontrol/osqp/blob/515ceba7c72c47ea73f01b136cb729d62ae27dc0/src/auxil.c#L887-L890
Is there a reason to not allow eps_abs positive and eps_rel zero? Many other QP solvers allow you to enforce an absolute tolerance on constraints, so setting eps_rel to zero would help when comparing OSQP to these solvers. Of course, it's possible to set it to an extremely tiny value, but you shouldn't need to do this.
The text was updated successfully, but these errors were encountered: