app.py

import gradio as gr
from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.chains import RetrievalQA
from langchain.llms import OpenAI

# Initialize the FAISS vector store
vector_store = None

# Function to handle PDF upload and indexing
def index_pdf(pdf):
    global vector_store
    
    # Load the PDF
    loader = PyPDFLoader(pdf.name)
    documents = loader.load()
    
    # Split the documents into chunks
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
    texts = text_splitter.split_documents(documents)
    
    # Embed the chunks and store them in the vector store
    embeddings = OpenAIEmbeddings()
    vector_store = FAISS.from_documents(texts, embeddings)
    
    return "PDF indexed successfully!"

# Function to handle chatbot queries
def chatbot_query(query):
    if vector_store is None:
        return "Please upload and index a PDF first."
    
    # Create a retrieval-based QA chain
    retriever = vector_store.as_retriever()
    qa_chain = RetrievalQA(llm=OpenAI(), retriever=retriever)
    
    # Get the response from the QA chain
    response = qa_chain.run(query)
    
    return response

# Create the Gradio interface
with gr.Blocks() as demo:
    with gr.Tab("Indexing"):
        pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
        index_button = gr.Button("Index PDF")
        index_output = gr.Textbox(label="Indexing Status")
        
        index_button.click(index_pdf, inputs=pdf_input, outputs=index_output)
    
    with gr.Tab("Chatbot"):
        query_input = gr.Textbox(label="Enter your question")
        query_button = gr.Button("Submit")
        query_output = gr.Textbox(label="Response")
        
        query_button.click(chatbot_query, inputs=query_input, outputs=query_output)

# Launch the Gradio app
demo.launch()