vector_store.py

import faiss
import numpy as np
from langchain.vectorstores import FAISS
from langchain.schema import Document
from utils.embeddings import embedding_model

# Initialize FAISS index
dim = 384  # MiniLM embedding size
index = faiss.IndexFlatL2(dim)
vector_db = FAISS(embedding_function=embedding_model, index=index)

def add_document(text, doc_id):
    """Add document embeddings to FAISS."""
    doc = Document(page_content=text, metadata={"id": doc_id})
    vector_db.add_documents([doc])

def search(query, top_k=3):
    """Search FAISS index for similar documents."""
    results = vector_db.similarity_search(query, k=top_k)
    return results