vector_store.py

import streamlit as st
from langchain_huggingface import HuggingFaceEmbeddings
from langchain_chroma import Chroma
from langchain_community.document_loaders import PyPDFLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter


@st.cache_resource()
def load_embedding_model(model):
    """
    sentence-transformers/all-mpnet-base-v2
    sentence-transformers/all-MiniLM-L6-v2
    """
    model = HuggingFaceEmbeddings(model_name=model)
    return model


def load_vector_store():
    """
    Loads a simple vector store
    I didn't use @st.cache because I want to
    load vector store on every page load
    """
    model = load_embedding_model("sentence-transformers/all-MiniLM-L6-v2")
    vector_store = Chroma(
        collection_name="main_store",
        embedding_function=model,
    )
    return vector_store


def process_pdf(pdf, vector_store):
    """
    Loads a pdf and splits it into chunks
    """
    loader = PyPDFLoader(pdf)
    docs = loader.load()
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
    splits = text_splitter.split_documents(docs)
    vector_store.add_documents(splits)