MaxDivProblem

MaxDivProblem dataclass

MaxDivProblem(
    vectors: NDArray[float32],
    k: int,
    distance_metric: DistanceMetric,
    diversity_metric: DiversityMetric,
    constraints: list[Constraint],
)

Immutable definition of a Maximum Diversity Problem.

A problem consists of n vectors in d dimensions, a target selection size k, a distance metric, a diversity metric, and optionally a list of fairness constraints.

Use the new factory method to create instances with validation.

new classmethod

new(
    vectors: ndarray,
    k: int,
    distance_metric: DistanceMetric = L2_EUCLIDEAN,
    diversity_metric: DiversityMetric = GEOMEAN_SEPARATION,
    constraints: list[Constraint] | None = None,
) -> Self

Create a new MaxDivProblem with validation.

Parameters:

Name	Type	Description	Default
vectors	ndarray	2D numpy array of shape (n, d) with at least 3 rows. Converted to float32 automatically if needed.	required
k	int	Number of vectors to select (must satisfy 2 <= k <= n).	required
distance_metric	DistanceMetric	Distance metric for pairwise distances.	L2_EUCLIDEAN
diversity_metric	DiversityMetric	Diversity metric to maximize.	GEOMEAN_SEPARATION
constraints	list[Constraint] | None	Optional list of fairness constraints.	None