app.py

from functions import *

# set the title
st.sidebar.title(DASHBOARD_TITLE)
info_section = st.empty()

# add an explanation of what is NER and why it is important for medical tasks
st.sidebar.markdown(
    f"""
    Meta Llama 3 8B Instruct is part of a family of large language models (LLMs) optimized for dialogue tasks.

    This project uses Streamlit to create a simple chatbot interface that allows you to chat with the model using the Hugging Face Inference API.
    
    Ask the model marketing-related questions and see how it responds. Have fun!
    
    Model used: [{MODEL_PATH}]({MODEL_LINK})
    """
)

first_assistant_message = "Hello! I am Marketing expert. What can I help you with today?"

# clear conversation
if st.sidebar.button("Clear conversation"):
    chat_history = [{'role':'assistant', 'content':first_assistant_message}]
    st.session_state['chat_history'] = chat_history
    st.rerun()

# Get the chat history
if "chat_history" not in st.session_state:
    chat_history = [{'role':'assistant', 'content':first_assistant_message}]
    st.session_state['chat_history'] = chat_history
else:
    chat_history = st.session_state['chat_history']

# print the conversation
for message in chat_history:
    with st.chat_message(message['role']):
        st.write(message['content'])

# keep only last 50 messages
short_history = [message for message in chat_history[-50:] if 'content' in message]

# include a system prompt to explain the bot what to do
short_history = [{'role': 'system', 'content': SYSTEM_PROMPT}] + short_history

# get the input from user
user_input = st.chat_input("Write something...")

if user_input:

    with st.chat_message("user"):
        st.write(user_input)

    # make the request
    with st.spinner("Generating the response..."):
        
        # create a shorter_history to avoid to keep a fair usage of the API
        short_history = short_history + [{'role': 'user', 'content': user_input}]

        # get the fill history for the next iteration
        chat_history = make_request(user_input, 
                                    short_history, 
                                    chat_history)

    st.session_state['chat_history'] = chat_history
    st.rerun()