import streamlit as st
from transformers import pipeline

# configuring streamlit page settings
st.set_page_config(
    page_title='Digital Ink',
    layout = 'centered'
)

generation_args = { 
    "max_new_tokens": 1000, 
    "return_full_text": False, 
    "num_beams": 5,
    "do_sample": True,
    "top_k": 60,
}

# Initialize the model pipeline
chat_pipeline = pipeline("text-generation", model="microsoft/Phi-3-mini-128k-instruct")


# Streamlit app
st.title("Digital Ink")

# Initialize the chat history
if 'chat_history' not in st.session_state:
    st.session_state.chat_history = []

#display chat history
for message in st.session_state.chat_history:
    with st.chat_message(message["role"]):
        st.markdown(message["content"])



# User input
user_input = st.chat_input("Ask Digital Ink..")

if user_input:
    # Add user message to chat history
    st.session_state.message.append({"role": "system", "content": "You are a helpful assistant named Digital Ink. Your purpose is to provide creative engaging and effective marketing content.You can introduce your self as follows: I'm Digital Ink, a marketing content generation model. I'm designed to assist you in creating engaging and effective marketing content, such as blog posts, social media posts, and product descriptions"})
    st.session_state.message.append({"role": "user", "content": user_input})
    st.chat_state.chat_message("user").markdown(user_input)

    # Generate response from chatbot
    context = [msg['content'] for msg in st.session_state.messages]
    message = [ 
    {"role": "system", "content": "You are a helpful assistant named Digital Ink. Your purpose is to provide creative engaging and effective marketing content.You can introduce your self as follows: I'm Digital Ink, a marketing content generation model. I'm designed to assist you in creating engaging and effective marketing content, such as blog posts, social media posts, and product descriptions"}, 
    {"role": "user", "content": user_input}, 
    {"role": "assistant", "content": ""}, 
    {"role": "user", "content": ""}, 
]
    response = chat_pipeline(message, **generation_args)[0]['generated_text']

    # Add assistant response to chat history
    st.session_state.messages.append({"role": "assistant", "content": response})
    with st.chat_message("assistant"):
        st.markdown(response)