src/streamlit_app.py

import streamlit as st
from langchain.document_loaders import TextLoader
from langchain.document_loaders import UnstructuredFileLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
from langchain.chains import RetrievalQA
from langchain.llms import HuggingFaceHub
import tempfile
import os

@st.cache_resource
def load_vector_store(file_path):
    # Load and chunk the document
    loader = TextLoader(file_path)
    documents = loader.load()

    splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
    chunks = splitter.split_documents(documents)

    # Create embeddings and store in FAISS
    embedding_model = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
    db = FAISS.from_documents(chunks, embedding_model)
    return db

def main():
    st.title("📄 Ask Questions About Your Document")
    st.write("Upload a `.txt` file and ask anything!")

    uploaded_file = st.file_uploader("Upload a text file", type=["txt"])

    if uploaded_file:
        with tempfile.NamedTemporaryFile(delete=False, suffix=".txt") as tmp_file:
            tmp_file.write(uploaded_file.read())
            tmp_path = tmp_file.name

        db = load_vector_store(tmp_path)

        query = st.text_input("Enter your question:")
        if query:
            qa_chain = RetrievalQA.from_chain_type(
                llm=HuggingFaceHub(
                    repo_id="google/flan-t5-base",
                    model_kwargs={"temperature": 0.5, "max_length": 256}
                ),
                retriever=db.as_retriever(),
                return_source_documents=True
            )

            result = qa_chain.run(query)

            st.write("### 📌 Answer")
            st.write(result)

        # Clean up temp file
        os.remove(tmp_path)

if __name__ == "__main__":
    main()