@astermind/astermind-synth/dist/encoders/CharVocab.d.ts

OmegaSynth - Label-Conditioned Synthetic Data Generator for AsterMind ELM/KELM Pipelines

/**
 * CharVocab - Character vocabulary builder
 * Builds a vocabulary from character sets and training data
 */
export declare class CharVocab {
    private charToIndex;
    private indexToChar;
    private size;
    /**
     * Build vocabulary from a set of strings
     * @param samples Array of strings to build vocabulary from
     * @param charSet Optional predefined character set (e.g., alphanumeric + punctuation)
     */
    build(samples: string[], charSet?: string): void;
    /**
     * Get index for a character
     */
    getIndex(char: string): number;
    /**
     * Get character for an index
     */
    getChar(index: number): string;
    /**
     * Check if character exists in vocabulary
     */
    hasChar(char: string): boolean;
    /**
     * Get vocabulary size
     */
    getSize(): number;
    /**
     * Get all characters in vocabulary
     */
    getChars(): string[];
    /**
     * Get default character set (alphanumeric + common punctuation)
     */
    static getDefaultCharSet(): string;
}