core/embeddings.py

from sentence_transformers import SentenceTransformer
from typing import List, Union
import numpy as np

class Embedder:
    def __init__(self, model_name: str = "all-MiniLM-L6-v2"):
        # Nutzt verfügbare Hardware (Cuda/CPU) automatisch
        self.model = SentenceTransformer(model_name)

    def encode(self, texts: Union[str, List[str]], normalize: bool = True) -> np.ndarray:
        """
        Encodes text(s) into embeddings.

        Args:
            texts: single string or list of strings
            normalize: optional normalization for cosine similarity use

        Returns:
            numpy.ndarray embeddings
        """
        if not texts:
            return np.array([])

        if isinstance(texts, str):
            if not texts.strip():
                return np.array([])
            texts = [texts]

        # Konvertiert leere Strings in der Liste, um Crashs zu vermeiden
        texts = [t if t.strip() else " " for t in texts]

        embeddings = self.model.encode(texts, normalize_embeddings=normalize)
        return embeddings