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
I am trying to implement an ILP solver based on ASTLogicSolver. However, I realize the Tnorm transformation generates non-linear constraints by using * to implement and and or. It should be possible to implement these ops in linear form, such that these constraints can be used in ILP.
The text was updated successfully, but these errors were encountered:
The default behavior treats operant as torch.tensor. Should I override the solver (and probably the whole set of operators) to support another operant type?
I am trying to implement an ILP solver based on
ASTLogicSolver
. However, I realize the Tnorm transformation generates non-linear constraints by using*
to implementand
andor
. It should be possible to implement these ops in linear form, such that these constraints can be used in ILP.The text was updated successfully, but these errors were encountered: