import os import requests import streamlit as st from langchain_huggingface import HuggingFaceEndpoint from langchain_core.prompts import PromptTemplate from langchain_core.output_parsers import StrOutputParser from transformers import pipeline from config import NASA_API_KEY # Ensure this file exists with your NASA API Key # Set up Streamlit UI st.set_page_config(page_title="HAL - NASA ChatBot", page_icon="🚀") # --- Ensure Session State Variables are Initialized --- if "chat_history" not in st.session_state: st.session_state.chat_history = [{"role": "assistant", "content": "Hello! How can I assist you today?"}] if "response_ready" not in st.session_state: st.session_state.response_ready = False # Tracks whether HAL has responded if "follow_up" not in st.session_state: st.session_state.follow_up = "" # Stores follow-up question # --- Set Up Model & API Functions --- model_id = "mistralai/Mistral-7B-Instruct-v0.3" # Initialize sentiment analysis pipeline with explicit model specification sentiment_analyzer = pipeline( "sentiment-analysis", model="distilbert/distilbert-base-uncased-finetuned-sst-2-english", revision="714eb0f" ) def get_llm_hf_inference(model_id=model_id, max_new_tokens=128, temperature=0.7): # Explicitly specify task="text-generation" so that the endpoint knows which task to run return HuggingFaceEndpoint( repo_id=model_id, max_new_tokens=max_new_tokens, temperature=temperature, token=os.getenv("HF_TOKEN"), task="text-generation" ) def get_nasa_apod(): url = f"https://api.nasa.gov/planetary/apod?api_key={NASA_API_KEY}" response = requests.get(url) if response.status_code == 200: data = response.json() return data.get("url", ""), data.get("title", ""), data.get("explanation", "") else: return "", "NASA Data Unavailable", "I couldn't fetch data from NASA right now. Please try again later." def analyze_sentiment(user_text): result = sentiment_analyzer(user_text)[0] return result['label'] def predict_action(user_text): if "NASA" in user_text or "space" in user_text: return "nasa_info" return "general_query" def generate_follow_up(user_text): """ Generates a concise and conversational follow-up question related to the user's input. """ prompt_text = ( f"Given the user's question: '{user_text}', generate a SHORT and SIMPLE follow-up question. " "Make it conversational and friendly. Example: 'Would you like to learn more about the six types of quarks?' " "Do NOT provide long explanations—just ask a friendly follow-up question." ) hf = get_llm_hf_inference(max_new_tokens=32, temperature=0.7) return hf.invoke(input=prompt_text).strip() def get_response(system_message, chat_history, user_text, max_new_tokens=256): """ Generates HAL's response, making it more conversational and engaging. """ sentiment = analyze_sentiment(user_text) action = predict_action(user_text) if action == "nasa_info": nasa_url, nasa_title, nasa_explanation = get_nasa_apod() response = f"**{nasa_title}**\n\n{nasa_explanation}" chat_history.append({'role': 'user', 'content': user_text}) chat_history.append({'role': 'assistant', 'content': response}) follow_up = generate_follow_up(user_text) chat_history.append({'role': 'assistant', 'content': follow_up}) return response, follow_up, chat_history, nasa_url hf = get_llm_hf_inference(max_new_tokens=max_new_tokens, temperature=0.9) prompt = PromptTemplate.from_template( ( "[INST] {system_message}\n\nCurrent Conversation:\n{chat_history}\n\nUser: {user_text}.\n [/INST]\n" "AI: Keep responses conversational and engaging. Start with a friendly phrase like " "'Certainly!', 'Of course!', or 'Great question!' before answering. " "Keep responses concise but engaging.\nHAL:" ) ) chat = prompt | hf.bind(skip_prompt=True) | StrOutputParser(output_key='content') response = chat.invoke(input=dict(system_message=system_message, user_text=user_text, chat_history=chat_history)) response = response.split("HAL:")[-1].strip() chat_history.append({'role': 'user', 'content': user_text}) chat_history.append({'role': 'assistant', 'content': response}) if sentiment == "NEGATIVE": response = "I'm here to help. Let me know what I can do for you. 😊" follow_up = generate_follow_up(user_text) chat_history.append({'role': 'assistant', 'content': follow_up}) return response, follow_up, chat_history, None # --- Chat UI --- st.title("🚀 HAL - Your NASA AI Assistant") st.markdown("🌌 *Ask me about space, NASA, and beyond!*") # Sidebar: Reset Chat if st.sidebar.button("Reset Chat"): st.session_state.chat_history = [{"role": "assistant", "content": "Hello! How can I assist you today?"}] st.session_state.response_ready = False st.session_state.follow_up = "" st.experimental_rerun() # Custom Chat Styling st.markdown(""" """, unsafe_allow_html=True) # Chat History Display st.markdown("