src/embedding_utils.py

import streamlit as st
from sentence_transformers import SentenceTransformer
import os
os.environ["SENTENCE_TRANSFORMERS_DISABLE_ONNX"] = "1"
os.environ["TRANSFORMERS_CACHE"] = "/tmp/.cache"
os.environ["HF_HOME"] = "/tmp/.cache"
os.makedirs("/tmp/.cache", exist_ok=True)

model = SentenceTransformer(
    "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
    cache_folder="/tmp/.cache"
)

def generate_embedding(text):

    if not text or not isinstance(text, str) or text.strip() == "":
        raise ValueError("Input text is empty or invalid.")
    return model.encode(text)