import streamlit as st
from transformers import pipeline

# Load the BioGPT model from HuggingFace or another medical GPT model
# BioGPT has been fine-tuned on medical data and should provide better responses
generator = pipeline("text-generation", model="microsoft/BioGPT")

# Streamlit app title
st.title("24/7Dr. Health Chatbot")

# Initialize session state for conversation history
if 'history' not in st.session_state:
    st.session_state.history = []

# Function to generate chatbot responses using BioGPT
def generate_medical_response(user_input):
    # Generate a response using BioGPT (or another medical model)
    response = generator(user_input,
                         max_length=150,
                         num_return_sequences=1,
                         pad_token_id=50256,
                         truncation=True,
                         temperature=0.7,
                         top_k=50,
                         top_p=0.95)
    return response[0]['generated_text']

# Input box for user symptoms
user_input = st.text_input("Describe your symptoms:")

if st.button("Ask"):
    if user_input:
        # Store the user's input in the conversation history
        st.session_state.history.append(f"You: {user_input}")

        # Generate the chatbot's response using the BioGPT model
        bot_response = generate_medical_response(user_input)

        # Store the chatbot's response in the conversation history
        st.session_state.history.append(f"Bot: {bot_response}")

        # Clear the input box
        user_input = ""

# Display the conversation history
if st.session_state.history:
    st.subheader("Conversation History")
    for message in st.session_state.history:
        st.write(message)