app.py

import streamlit as st
from streamlit_chat import message
# Initialize the CSVLoader to load the uploaded CSV file
from langchain.document_loaders.csv_loader import CSVLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
from langchain.llms import CTransformers
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
import sys

# Display the title of the web page
st.title("Chat with CSV using open source LLM Inference Point 🦙🦜")
# Display a markdown message with additional information
st.markdown("<h3 style='text-align: center; color: white;'>Built by <a href='https://github.com/AIAnytime'>AI Anytime with ❤️ </a></h3>", unsafe_allow_html=True)

# Allow users to upload a CSV file
uploaded_file = st.sidebar.file_uploader("Upload your Data", type="csv")

if uploaded_file :
# Initialize the CSVLoader to load the uploaded CSV file
   #use tempfile because CSVLoader only accepts a file_path
    with tempfile.NamedTemporaryFile(delete=False) as tmp_file:
        tmp_file.write(uploaded_file.getvalue())
        tmp_file_path = tmp_file.name

    db = DB_FAISS_PATH = "vectorstore/db_faiss"
# Initialize the CSVLoader to load the uploaded CSV file
    loader = CSVLoader(file_path="data/2019.csv", encoding="utf-8", csv_args={'delimiter': ','})
    data = loader.load()
    db.save_local(DB_FAISS_PATH)
    llm = load_llm()

    chain = ConversationalRetrievalChain.from_llm(llm=llm, retriever=db.as_retriever())

    def conversational_chat(query):
# Maintain and display the chat history
            result = chain({"question": query, "chat_history": st.session_state['history']})
# Maintain and display the chat history
            st.session_state['history'].append((query, result["answer"]))
            return result["answer"]
    
# Maintain and display the chat history
    if 'history' not in st.session_state:
# Maintain and display the chat history
        st.session_state['history'] = []

# Maintain and display the chat history
    if 'generated' not in st.session_state:
# Maintain and display the chat history
        st.session_state['generated'] = ["Hello ! Ask me anything about " + uploaded_file.name + " 🤗"]

# Maintain and display the chat history
    if 'past' not in st.session_state:
# Maintain and display the chat history
        st.session_state['past'] = ["Hey ! 👋"]
        
    #container for the chat history
    response_container = st.container()
    #container for the user's text input
    container = st.container()

    with container:
        with st.form(key='my_form', clear_on_submit=True):
            
            user_input = st.text_input("Query:", placeholder="Talk to your csv data here (:", key='input')
            submit_button = st.form_submit_button(label='Send')
            
        if submit_button and user_input:
            output = conversational_chat(user_input)
            
# Maintain and display the chat history
            st.session_state['past'].append(user_input)
# Maintain and display the chat history
            st.session_state['generated'].append(output)

# Maintain and display the chat history
    if st.session_state['generated']:
        with response_container:
# Maintain and display the chat history
            for i in range(len(st.session_state['generated'])):
# Maintain and display the chat history
                message(st.session_state["past"][i], is_user=True, key=str(i) + '_user', avatar_style="big-smile")
# Maintain and display the chat history
                message(st.session_state["generated"][i], key=str(i), avatar_style="thumbs")
            
# Split the text into Chunks
text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=20)
text_chunks = text_splitter.split_documents(data)

print(len(text_chunks))

# Download Sentence Transformers Embedding From Hugging Face
embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2',
                                       model_kwargs={'device': 'cpu'})

# COnverting the text Chunks into embeddings and saving the embeddings into FAISS Knowledge Base
docsearch = FAISS.from_documents(text_chunks, embeddings)

docsearch.save_local(DB_FAISS_PATH)


#query = "What is the value of GDP per capita of Finland provided in the data?"

#docs = docsearch.similarity_search(query, k=3)

#print("Result", docs)

from transformers import pipeline

pipe = pipeline("text-generation",model="mistralai/Mistral-7B-v0.1",model_type="llama",max_new_tokens=512,temperature=0.1 )

qa = ConversationalRetrievalChain.from_llm(llm, retriever=docsearch.as_retriever())

# Insert a chat message container.
with st.chat_message("user"):
    st.write("Hello 👋")
    st.line_chart(np.random.randn(30, 3))

# Display a chat input widget.
st.chat_input("Say something")

while True:
    chat_history = []
    #query = "What is the value of  GDP per capita of Finland provided in the data?"
    query = input(f"Input Prompt: ")
    if query == 'exit':
        print('Exiting')
        sys.exit()
    if query == '':
        continue
    result = qa({"question":query, "chat_history":chat_history})
    print("Response: ", result['answer'])