IntegerBounds

class ioh.iohcpp.IntegerBounds(self, lb: ndarray[shape=(*), order='C', device='cpu'], ub: ndarray[shape=(*), order='C', device='cpu'], enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT)

class ioh.iohcpp.IntegerBounds(self, lb: collections.abc.Sequence[int], ub: collections.abc.Sequence[int], enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT)

class ioh.iohcpp.IntegerBounds(self, size: int, lb: int, ub: int, enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT)

Bases: AbstractIntegerConstraint

Overloaded function.

1. __init__(self, lb: ndarray[shape=(*), order='C', device='cpu'], ub: ndarray[shape=(*), order='C', device='cpu'], enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT) -> None

2. __init__(self, lb: collections.abc.Sequence[int], ub: collections.abc.Sequence[int], enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT) -> None

   Problem Bounds, defined as a Box Constraint

   lb: list

   the lower bound of the search space/domain

   ub: list

   the upper bound of the search space/domain

   enforced: ConstraintEnforcement

   whether the constraint should be enforced

3. __init__(self, size: int, lb: int, ub: int, enforced: ioh.iohcpp.ConstraintEnforcement = ConstraintEnforcement.NOT) -> None

   Problem Bounds, defined as a Box Constraint

   size: int

   The size of the problem

   lb: float

   the lower bound of the search space/domain

   ub: float

   the upper bound of the search space/domain

   enforced: ConstraintEnforcement

   whether the constraint should be enforced

Attributes Summary

enforced	The type of constraint enforcement applied.
exponent	The exponent for this constraint
lb	a new numpy array on every read (assign lb to change it)
ub	a new numpy array on every read (assign ub to change it)
weight	The weight given to this constraint

Methods Summary

__call__([order, device])
compute_violation([order, device])	Overloaded function.
is_feasible([order, device])
penalize(self, arg, /)
penalty(self)
violation(self)

Attributes Documentation

enforced

The type of constraint enforcement applied.

exponent

The exponent for this constraint

lb

a new numpy array on every read (assign lb to change it)

Type:

The lower bound (box constraint)

ub

a new numpy array on every read (assign ub to change it)

Type:

The upper bound (box constraint)

weight

The weight given to this constraint

Methods Documentation

__call__(self, arg0: ndarray[shape=(*), order='C', device='cpu'], arg1: float, /) → float

__call__(self, arg0: collections.abc.Sequence[int], arg1: float, /) → float

compute_violation(self, arg: ndarray[shape=(*), order='C', device='cpu'], /) → bool

compute_violation(self, arg: collections.abc.Sequence[int], /) → bool

Overloaded function.

1. compute_violation(self, arg: ndarray[shape=(*), order='C', device='cpu'], /) -> bool

2. compute_violation(self, arg: collections.abc.Sequence[int], /) -> bool

   Check if a point is within the bounds or not.

   x: list | np.ndarray

   the search point to check

   y: float

   the objective value

   bool

   Whether there is a constraint violation

is_feasible(self, arg: ndarray[shape=(*), order='C', device='cpu'], /) → bool

is_feasible(self, arg: collections.abc.Sequence[int], /) → bool

penalize(self, arg: float, /) → float

penalty(self) → float

violation(self) → float