|
import streamlit as st |
|
from dotenv import load_dotenv |
|
from PyPDF2 import PdfReader |
|
from langchain.text_splitter import CharacterTextSplitter |
|
from langchain_community.embeddings import OpenAIEmbeddings, HuggingFaceInstructEmbeddings |
|
from langchain_community.vectorstores import FAISS |
|
from langchain_community.chat_models import ChatOpenAI |
|
|
|
from langchain.memory import ConversationBufferMemory |
|
from langchain.chains import ConversationalRetrievalChain |
|
from htmlTemplates import css, bot_template, user_template, hide_st_style, footer |
|
from langchain_community.llms import HuggingFaceHub |
|
from matplotlib import style |
|
|
|
|
|
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 = CharacterTextSplitter( |
|
separator="\n", |
|
chunk_size=1000, |
|
chunk_overlap=200, |
|
length_function=len |
|
) |
|
chunks = text_splitter.split_text(text) |
|
return chunks |
|
|
|
|
|
def get_vectorstore(text_chunks): |
|
|
|
print("HuggingFaceInstructEmbeddings") |
|
model_kwargs = {'device': 'cpu', 'weights_only': True} |
|
|
|
embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl") |
|
print("FAISS.from_texts") |
|
vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings) |
|
print("returning vectorstore") |
|
return vectorstore |
|
|
|
|
|
def get_conversation_chain(vectorstore): |
|
|
|
llm = HuggingFaceHub(repo_id="google/flan-t5-base", model_kwargs={"temperature":0.5, "max_length":512}) |
|
|
|
memory = ConversationBufferMemory( |
|
memory_key='chat_history', return_messages=True) |
|
conversation_chain = ConversationalRetrievalChain.from_llm( |
|
llm=llm, |
|
retriever=vectorstore.as_retriever(), |
|
memory=memory |
|
) |
|
return conversation_chain |
|
|
|
|
|
def handle_userinput(user_question): |
|
if st.session_state.conversation is None: |
|
st.error("Please upload PDF data before starting the chat.") |
|
return |
|
|
|
response = st.session_state.conversation({'question': user_question}) |
|
st.session_state.chat_history = response['chat_history'] |
|
|
|
for i, message in enumerate(st.session_state.chat_history): |
|
if i % 2 == 0: |
|
st.write(user_template.replace( |
|
"{{MSG}}", message.content), unsafe_allow_html=True) |
|
else: |
|
st.write(bot_template.replace( |
|
"{{MSG}}", message.content), unsafe_allow_html=True) |
|
|
|
|
|
def main(): |
|
load_dotenv() |
|
st.set_page_config(page_title="Talk with PDF", |
|
page_icon="icon.png") |
|
st.write(css, unsafe_allow_html=True) |
|
|
|
if "conversation" not in st.session_state: |
|
st.session_state.conversation = None |
|
if "chat_history" not in st.session_state: |
|
st.session_state.chat_history = None |
|
|
|
st.header("Chat with AI with Custom Data 🚀") |
|
user_question = st.text_input("Ask a question about your Data:") |
|
|
|
with st.sidebar: |
|
st.subheader("Your documents") |
|
pdf_docs = st.file_uploader( |
|
"Upload your Data here in PDF format and click on 'Process'", accept_multiple_files=True, type=['pdf']) |
|
|
|
if st.button("Process"): |
|
if pdf_docs is None: |
|
st.error("Please upload at least one PDF file.") |
|
else: |
|
with st.spinner("Processing"): |
|
print("get_pdf_text") |
|
raw_text = get_pdf_text(pdf_docs) |
|
print("get_text_chunks") |
|
text_chunks = get_text_chunks(raw_text) |
|
print("get_vectorstore") |
|
vectorstore = get_vectorstore(text_chunks) |
|
print("get_conversation_chain") |
|
st.session_state.conversation = get_conversation_chain( |
|
vectorstore) |
|
print("success") |
|
st.success("Your Data has been processed successfully") |
|
|
|
if user_question: |
|
handle_userinput(user_question) |
|
|
|
st.markdown(hide_st_style, unsafe_allow_html=True) |
|
st.markdown(footer, unsafe_allow_html=True) |
|
|
|
|
|
if __name__ == '__main__': |
|
main() |
|
|
