saealib.Comparator

class saealib.Comparator(weights, eps_cv, eps_obj, direction=None)

Bases: ABC

Base class for comparator.

Parameters:

• weights (np.ndarray)

• eps_cv (float)

• eps_obj (float)

• direction (np.ndarray | None)

weights

Weights for objectives. shape = (n_obj, )

Type:

np.ndarray

eps_cv

Epsilon for constraint violation feasibility threshold. Under Optimizer execution this value is overwritten every generation from problem.handler.feasibility_threshold; the constructor argument only matters for standalone (non-Optimizer) use.

Type:

float

eps_obj

Epsilon for objective value equality comparison.

Type:

float

direction

Per-objective optimization directions (+1 = maximize, -1 = minimize). None means all objectives are minimized.

Type:

np.ndarray or None

Methods

__init__
compare	Compare two solutions directly from raw values.
compare_population	Compare individual a and b in population.
get_required_attrs	Return population attributes this comparator needs to persist.
prepare_population	Freshly recompute and materialize population-relative ranking state.
rank_population	Prepare population-relative ranking state, then sort.
sort_population	Sort population based on their attributes.

Method Details

abstractmethod Comparator.__init__(weights, eps_cv, eps_obj, direction=None)

Parameters:

• weights (ndarray)

• eps_cv (float)

• eps_obj (float)

• direction (ndarray | None)

abstractmethod Comparator.compare(fa, cv_a, fb, cv_b)

Compare two solutions directly from raw values.

This is the low-overhead variant of compare_population that avoids constructing or indexing into a Population object.

Parameters:

• fa (np.ndarray) – Objective values of solution a. shape = (n_obj,)

• cv_a (float) – Constraint violation of solution a.

• fb (np.ndarray) – Objective values of solution b. shape = (n_obj,)

• cv_b (float) – Constraint violation of solution b.

Returns:

-1 if a is better than b, 1 if b is better than a, 0 if equal.

Return type:

int

abstractmethod Comparator.compare_population(population, idx_a, idx_b)

Compare individual a and b in population.

Parameters:

• population (Population) – The population to compare.

• idx_a (int) – Index of the first individual.

• idx_b (int) – Index of the second individual.

Returns:

Comparison result. -1 if the individual a is better than the individual b. 1 if the individual a is worse than the individual b. 0 if the individual a is equal to the individual b.

Return type:

int

Comparator.get_required_attrs(problem)

Return population attributes this comparator needs to persist.

Population-relative ranking state (e.g. SPEA2 fitness) that must survive across selections belongs here as a named, typed attribute rather than in the population’s transient cache. Default: no additional attributes.

Parameters:

problem (Problem) – The optimization problem.

Returns:

Empty by default.

Return type:

list[PopulationAttribute]

Comparator.prepare_population(population)

Freshly recompute and materialize population-relative ranking state.

Default is a no-op — comparators whose ranking depends only on per-individual values already stored in f/cv do not need this. Comparators that declare attributes via get_required_attrs() should override this to recompute and write those attributes for the given population, unconditionally (not reusing any prior cached value).

Parameters:

population (Population) – The population to prepare ranking state for.

Return type:

None

Comparator.rank_population(population)

Prepare population-relative ranking state, then sort.

Equivalent to calling prepare_population() followed by sort_population(). Callers performing environmental selection on a freshly assembled pool (e.g. survivor selection merging population and offspring) should call this instead of sort_population() directly, so any population-relative state declared via get_required_attrs() is recomputed for the new set before ranking.

Parameters:

population (Population) – The population to rank.

Returns:

Sorted population indices, as returned by sort_population().

Return type:

np.ndarray

abstractmethod Comparator.sort_population(population)

Sort population based on their attributes.

Parameters:

population (Population) – The population to sort.

Returns:

Sorted population indices.

Return type:

np.ndarray