app.py

from fastapi import FastAPI, HTTPException
import faiss
import numpy as np
from sentence_transformers import SentenceTransformer
from datasets import load_dataset

app = FastAPI()

# 1. 임베딩 모델 로드
embed_model = SentenceTransformer("all-MiniLM-L6-v2")

# 2. Hugging Face에서 MedRAG 데이터셋 로드
dataset = load_dataset("MedRAG/textbooks", split="train", streaming=True)

# 3. 데이터 변환
texts = [entry["content"] for entry in dataset]  # "content" 필드 활용