embeddings/embedding_model.py

from __future__ import annotations

import logging
from functools import lru_cache
from typing import List

from sentence_transformers import SentenceTransformer

logger = logging.getLogger(__name__)


@lru_cache(maxsize=1)
def get_embedding_model(model_name: str = "sentence-transformers/all-MiniLM-L6-v2") -> SentenceTransformer:
    """

    Return a cached SentenceTransformers model instance.



    Note: loading the model can be slow; caching keeps Streamlit responsive.

    """

    logger.info("Loading embedding model: %s", model_name)
    return SentenceTransformer(model_name)


def embed_texts(texts: List[str], model_name: str = "sentence-transformers/all-MiniLM-L6-v2") -> List[List[float]]:
    model = get_embedding_model(model_name=model_name)
    vectors = model.encode(texts, normalize_embeddings=True)
    return [v.tolist() for v in vectors]