app.py

import streamlit as st
import pdfplumber
import base64
from langchain.llms import OpenAI
from langchain.vectorstores.cassandra import Cassandra
from langchain.indexes.vectorstore import VectorStoreIndexWrapper
from langchain.embeddings import OpenAIEmbeddings
from typing_extensions import Concatenate
from datasets import load_dataset
from langchain.memory import ConversationBufferWindowMemory
import cassio

from PyPDF2 import PdfReader


def main():
    st.title("INTERACTION WITH PDF USING LLM")

    pdf_file = st.file_uploader("Upload PDF file", type=["pdf"])

    if pdf_file is not None:
 
        ASTRA_DB_APPLICATION_TOKEN="AstraCS:KRrILGTZHQMczBfoJhucdxkN:a6aaf66c8f7e318f1048bb13ec9132510c3fefc85501a5268cd873edd418ad10"
        ASTRA_DB_ID="800e9596-9d6a-487d-a87c-b95436d8026a"
        OPENAI_API_KEY="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"
        
        pdfreader=PdfReader(pdf_file)
  
        raw_text=''
        
        for i ,page in enumerate(pdfreader.pages):
          content=page.extract_text()
          if content:
            raw_text += content
 
        cassio.init(token=ASTRA_DB_APPLICATION_TOKEN,database_id=ASTRA_DB_ID)

        
        
        llm=OpenAI(openai_api_key=OPENAI_API_KEY)
        
        embedding=OpenAIEmbeddings(openai_api_key=OPENAI_API_KEY)
        
        astra_vector_store=Cassandra(embedding=embedding,
                                     table_name='qa_mini_demo',
                                     session=None,
                                     keyspace=None,
                                     )
        
        astra_vector_store.delete_collection()
        
        from langchain.text_splitter import CharacterTextSplitter
        text_splitter=CharacterTextSplitter(
            separator='\n',
            chunk_size=800,
            chunk_overlap=200,
            length_function=len
        )
        
        texts=text_splitter.split_text(raw_text)
        
        
        astra_vector_store.add_texts(texts)
        
        astra_vector_index=VectorStoreIndexWrapper(vectorstore=astra_vector_store)
        
        
        query_text = st.text_input("Enter your Question:").strip()
        
        submit=st.button('Generate')

        
        if submit:

            answer = astra_vector_index.query(query_text, llm=llm).strip()
        
            st.write("\nANSWER :\"%s\"" % answer)
        
    
if __name__ == "__main__":
    main()