export declare class EmbeddingGenerator {
    private modelName;
    private dimension;
    constructor(modelName?: string, dimension?: number);
    initialize(): Promise<void>;
    generateEmbedding(text: string): Promise<number[]>;
    private simpleHash;
    generateBatchEmbeddings(texts: string[]): Promise<number[][]>;
    getDimension(): number;
    getModelName(): string;
    isInitialized(): boolean;
    static cosineSimilarity(a: number[], b: number[]): number;
    static euclideanDistance(a: number[], b: number[]): number;
    static normalizeVector(vector: number[]): number[];
    static addVectors(a: number[], b: number[]): number[];
    static subtractVectors(a: number[], b: number[]): number[];
    static scaleVector(vector: number[], scale: number): number[];
    static averageVectors(vectors: number[][]): number[];
    static preprocessText(text: string): string;
    static chunkText(text: string, maxLength?: number, overlap?: number): string[];
    semanticSearch(query: string, documents: string[], topK?: number): Promise<Array<{
        text: string;
        score: number;
        index: number;
    }>>;
    static kMeansClustering(embeddings: number[][], k: number, maxIterations?: number): {
        clusters: number[][][];
        centroids: number[][];
        assignments: number[];
    };
}
//# sourceMappingURL=embedding-generator.d.ts.map