import streamlit as st from PyPDF2 import PdfReader from langchain.text_splitter import RecursiveCharacterTextSplitter import os from langchain_google_genai import GoogleGenerativeAIEmbeddings import google.generativeai as genai from langchain.vectorstores import FAISS from langchain_google_genai import ChatGoogleGenerativeAI from langchain.chains.question_answering import load_qa_chain from langchain.prompts import PromptTemplate from dotenv import load_dotenv load_dotenv() os.getenv("GOOGLE_API_KEY") genai.configure(api_key=os.getenv("GOOGLE_API_KEY")) def get_pdf_text(pdf_docs): text="" for pdf in pdf_docs: pdf_reader= PdfReader(pdf) for page in pdf_reader.pages: text+= page.extract_text() return text def get_text_chunks(text): text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200) chunks = text_splitter.split_text(text) return chunks def get_vector_store(text_chunks): embeddings = GoogleGenerativeAIEmbeddings(model = "models/embedding-001") vector_store = FAISS.from_texts(text_chunks, embedding=embeddings) vector_store.save_local("faiss_index") def get_conversational_chain(): prompt_template = """ Use the following pieces of information to answer the user's question.\n\n Context: answer as long and as detailed as you can. Make specific points. {context}? Question: {question}? You are a helper chatbot. You answer people's questions. You have knowledge about everything in general. If you can't find information in the PDF, use your own knowledge to answer questions that are indirectly related to the PDF. However, make sure to connect your answers to the PDF's content, even when using external knowledge. Try your best to give the answer. Also try to add some your own wordings the describe the answer. Never Answer Like : "I don't know" , "The provided document does not contain information", "Bu sorunun cevabı verilen metinde bulunmamaktadır", "Metinde .... ilgili herhangi bir bilgi verilmemiştir.", ".... hakkında bilgi verilmemiştir.", Helpful Answer: """ model = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.4) prompt = PromptTemplate(template = prompt_template, input_variables = ["context", "question"]) chain = load_qa_chain(model, chain_type="stuff", prompt=prompt) return chain def user_input(user_question): embeddings = GoogleGenerativeAIEmbeddings(model = "models/embedding-001") new_db = FAISS.load_local("faiss_index", embeddings) docs = new_db.similarity_search(user_question) chain = get_conversational_chain() response = chain( {"input_documents":docs, "question": user_question} , return_only_outputs=True) print(response) st.write("Reply: ", response["output_text"]) def main(): st.set_page_config("Chat PDF") st.header("Chat with PDF using Gemini!") user_question = st.text_input("Ask a Question from the PDF Files") if user_question: user_input(user_question) with st.sidebar: st.audio("music.mp3", format='audio/mp3') st.image("img.jpg") st.write("---") pdf_docs = st.file_uploader("Upload your PDF Files and Click on the Submit & Process Button", accept_multiple_files=True) if st.button("Submit & Process"): with st.spinner("Processing..."): raw_text = get_pdf_text(pdf_docs) text_chunks = get_text_chunks(raw_text) get_vector_store(text_chunks) st.success("Done") if __name__ == "__main__": main()