Crate pare 

Pareto frontier and skyline query primitives for multi-objective optimization.

Maintains a set of non-dominated points across multiple dimensions, supporting maximization/minimization per dimension and crowding distance for diversity maintenance.

Modules

sensitivity

Objective redundancy analysis via the gradient test.

Structs

EpsilonArchive

Grid-based epsilon-dominance archive with bounded size.

NormalizationStats

Statistics for normalizing objective values.

ParetoFrontier

Pareto frontier maintaining a set of non-dominated points.

Point

A point in multi-objective space.

Enums

Direction

Direction of optimization for a dimension.

FrontierError

Errors returned when constructing a frontier from raw vectors.

Functions

dominates

Standalone dominance check.

generational_distance

Generational Distance (GD) from a front to a reference set.

inverted_generational_distance

Inverted Generational Distance (IGD) from a front to a reference set.

pareto_indices

Return indices of the Pareto frontier (all objectives maximized).

pareto_indices_2d

Return indices of the Pareto frontier using a simple 2D sweep (maximize/maximize).

pareto_indices_k_dominance

Return indices of the k-dominant frontier (maximize/maximize/…).

pareto_layers

Partition points into successive non-dominated layers (Pareto layers).