/**
 * The meta data for the weights
 */
export type WeightsMeta = {
    /**
     * The id of the weights
     */
    id?: string;
    /**
     * The type of the weights
     */
    type?: 'knn' | 'threshold' | 'queen' | 'rook' | 'kernel';
    /**
     * The symmetry of the weights
     */
    symmetry?: 'symmetric' | 'asymmetric';
    /**
     * The number of observations
     */
    numberOfObservations: number;
    /**
     * The minimum number of neighbors
     */
    minNeighbors: number;
    /**
     * The maximum number of neighbors
     */
    maxNeighbors: number;
    /**
     * The mean number of neighbors
     */
    meanNeighbors: number;
    /**
     * The median number of neighbors
     */
    medianNeighbors: number;
    /**
     * The percentage of non-zero neighbors
     */
    pctNoneZero: number;
    /**
     * The order of the weights
     */
    order?: number;
    /**
     * Whether to include lower order neighbors
     */
    includeLowerOrder?: boolean;
    /**
     * The k value for k-nearest neighbors
     */
    k?: number;
    /**
     * The threshold for the weights
     */
    threshold?: number;
    /**
     * The bandwidth for kernel weights
     */
    bandwidth?: number;
    /**
     * The kernel function for kernel weights
     */
    kernel?: string;
    /**
     * The power (or exponent) applied to the distance before normalizing by the bandwidth
     */
    power?: number;
    /**
     * Whether the distance is in miles
     */
    isMile?: boolean;
    /**
     * The distance metric for the weights
     */
    distanceMetric?: 'euclidean' | 'manhattan' | 'arc' | 'projected';
    /**
     * The unit of the distance metric
     */
    distanceUnit?: 'Foot_US' | 'Yard_US' | 'Meter' | 'Kilometer' | 'NauticalMile' | 'Degree' | 'Radian';
};
/**
 * Get the meta data from the weights structure
 * @param weights the weights structure of every observation using row index
 * @returns WeightsMeta
 */
export declare function getMetaFromWeights(weights: number[][], isDistanceWeights?: boolean): WeightsMeta;