Update app.py

app.py

@@ -4,191 +4,162 @@ 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
+from transformers import pipeline  # for Sentiment Analysis
+from config import NASA_API_KEY  # Import the NASA API key from the configuration file
 
-# 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"
 
-# Explicitly define the sentiment analysis model
-sentiment_analyzer = pipeline("sentiment-analysis", model="distilbert/distilbert-base-uncased-finetuned-sst-2-english", revision="714eb0f")
+# Initialize sentiment analysis pipeline
+sentiment_analyzer = pipeline("sentiment-analysis")
 
 def get_llm_hf_inference(model_id=model_id, max_new_tokens=128, temperature=0.1):
-    """
-    Initializes the Hugging Face text generation model with correct settings.
-    """
-    return HuggingFaceEndpoint(
+    llm = HuggingFaceEndpoint(
         repo_id=model_id,
-        task="text-generation",  # Explicitly define the task
         max_new_tokens=max_new_tokens,
         temperature=temperature,
-        token=os.getenv("HF_TOKEN")  # Ensure your Hugging Face API Token is set
+        token=os.getenv("HF_TOKEN")  # Hugging Face token from environment variable
     )
+    return llm
 
 def get_nasa_apod():
+    """
+    Fetch the Astronomy Picture of the Day (APOD) from the NASA API.
+    """
     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", "")
+        return f"Title: {data['title']}\nExplanation: {data['explanation']}\nURL: {data['url']}"
     else:
-        return "", "NASA Data Unavailable", "I couldn't fetch data from NASA right now. Please try again later."
+        return "I couldn't fetch data from NASA right now. Please try again later."
 
 def analyze_sentiment(user_text):
+    """
+    Analyzes the sentiment of the user's input to adjust responses.
+    """
     result = sentiment_analyzer(user_text)[0]
-    return result['label']
+    sentiment = result['label']
+    return sentiment
 
 def predict_action(user_text):
+    """
+    Predicts actions based on user input (e.g., fetch space info or general knowledge).
+    """
     if "NASA" in user_text or "space" in user_text:
         return "nasa_info"
+    if "weather" in user_text:
+        return "weather_info"
     return "general_query"
 
 def generate_follow_up(user_text):
     """
-    Generates a concise and conversational follow-up question related to the user's input.
+    Generates a relevant follow-up question based on 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."
+        f"Given the user's message: '{user_text}', ask one natural follow-up question "
+        "that suggests a related topic or offers user the opportunity to go in a new direction."
     )
 
-    hf = get_llm_hf_inference(max_new_tokens=32, temperature=0.7)
-    return hf.invoke(input=prompt_text).strip()
+    hf = get_llm_hf_inference(max_new_tokens=64, temperature=0.7)
+    chat = hf.invoke(input=prompt_text)
+    
+    return chat.strip()
 
-def get_response(system_message, chat_history, user_text, max_new_tokens=256):
-    """
-    Generates HAL's response, making it more conversational and engaging.
-    """
+def get_response(system_message, chat_history, user_text, 
+                 eos_token_id=['User'], max_new_tokens=256, get_llm_hf_kws={}):
     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}"
+        nasa_response = get_nasa_apod()
         chat_history.append({'role': 'user', 'content': user_text})
-        chat_history.append({'role': 'assistant', 'content': response})
+        chat_history.append({'role': 'assistant', 'content': nasa_response})
         
         follow_up = generate_follow_up(user_text)
         chat_history.append({'role': 'assistant', 'content': follow_up})
-        return response, follow_up, chat_history, nasa_url
+        return f"{nasa_response}\n\n{follow_up}", chat_history
 
-    hf = get_llm_hf_inference(max_new_tokens=max_new_tokens, temperature=0.9)
+    hf = get_llm_hf_inference(max_new_tokens=max_new_tokens, temperature=0.1)
 
     prompt = PromptTemplate.from_template(
         (
             "[INST] {system_message}"
             "\nCurrent Conversation:\n{chat_history}\n\n"
             "\nUser: {user_text}.\n [/INST]"
-            "\nAI: 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:"
+            "\nAI:"
         )
     )
-
     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()
+    response = response.split("AI:")[-1]
 
     chat_history.append({'role': 'user', 'content': user_text})
     chat_history.append({'role': 'assistant', 'content': response})
 
+    # Modify response based on sentiment analysis (e.g., offer help for negative sentiments)
     if sentiment == "NEGATIVE":
-        response = "I'm here to help. Let me know what I can do for you. 😊"
+        response += "\nI'm sorry to hear that. How can I assist you further?"
 
     follow_up = generate_follow_up(user_text)
     chat_history.append({'role': 'assistant', 'content': follow_up})
 
-    return response, follow_up, chat_history, None
+    return f"{response}\n\n{follow_up}", chat_history
 
-# --- Chat UI ---
-st.title("🚀 HAL - Your NASA AI Assistant")
-st.markdown("🌌 *Ask me about space, NASA, and beyond!*")
+# Streamlit setup
+st.set_page_config(page_title="HuggingFace ChatBot", page_icon="🤗")
+st.title("NASA Personal Assistant")
+st.markdown(f"*This chatbot uses {model_id} and NASA's APIs to provide information and responses.*")
 
-# Sidebar: Reset Chat
+# Initialize session state
+if "chat_history" not in st.session_state:
+    st.session_state.chat_history = [{"role": "assistant", "content": "Hello! How can I assist you today?"}]
+    
+# Sidebar for settings
 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("""
-    <style>
-    .user-msg {
-        background-color: #0078D7;
-        color: white;
-        padding: 10px;
-        border-radius: 10px;
-        margin-bottom: 5px;
-        width: fit-content;
-        max-width: 80%;
-    }
-    .assistant-msg {
-        background-color: #333333;
-        color: white;
-        padding: 10px;
-        border-radius: 10px;
-        margin-bottom: 5px;
-        width: fit-content;
-        max-width: 80%;
-    }
-    .container {
-        display: flex;
-        flex-direction: column;
-        align-items: flex-start;
-    }
-    @media (max-width: 600px) {
-        .user-msg, .assistant-msg { font-size: 16px; max-width: 100%; }
-    }
-    </style>
-""", unsafe_allow_html=True)
-
-# Chat History Display
-st.markdown("<div class='container'>", unsafe_allow_html=True)
-for message in st.session_state.chat_history:
-    if message["role"] == "user":
-        st.markdown(f"<div class='user-msg'><strong>You:</strong> {message['content']}</div>", unsafe_allow_html=True)
-    else:
-        st.markdown(f"<div class='assistant-msg'><strong>HAL:</strong> {message['content']}</div>", unsafe_allow_html=True)
-st.markdown("</div>", unsafe_allow_html=True)
-
-# --- Single Input Box for Both Initial and Follow-Up Messages ---
-user_input = st.chat_input("Type your message here...")  # Only ONE chat_input()
 
+# Main chat interface
+user_input = st.chat_input(placeholder="Type your message here...")
 if user_input:
-    response, follow_up, st.session_state.chat_history, image_url = get_response(
+    response, st.session_state.chat_history = get_response(
         system_message="You are a helpful AI assistant.",
         user_text=user_input,
-        chat_history=st.session_state.chat_history
+        chat_history=st.session_state.chat_history,
+        max_new_tokens=128
     )
+    # Display messages
+    for message in st.session_state.chat_history:
+        st.chat_message(message["role"]).write(message["content"])
+
+
+
+
+if st.button("Send"):
+    if user_input:
+        response, follow_up, st.session_state.chat_history, image_url = get_response(
+            system_message="You are a helpful AI assistant.",
+            user_text=user_input,
+            chat_history=st.session_state.chat_history
+        )
+
+        # Display response
+        st.markdown(f"<div class='assistant-msg'><strong>HAL:</strong> {response}</div>", unsafe_allow_html=True)
 
-    st.markdown(f"<div class='assistant-msg'><strong>HAL:</strong> {response}</div>", unsafe_allow_html=True)
+        # Display NASA image if available
+        if image_url:
+            st.image(image_url, caption="NASA Image of the Day")
 
-    if image_url:
-        st.image(image_url, caption="NASA Image of the Day")
+        # Follow-up question suggestions
+        follow_up_options = [follow_up, "Explain differently", "Give me an example"]
+        selected_option = st.radio("What would you like to do next?", follow_up_options)
 
-    st.session_state.follow_up = follow_up
-    st.session_state.response_ready = True  # Enables follow-up response cycle
+        if st.button("Continue"):
+            if selected_option:
+                response, _, st.session_state.chat_history, _ = get_response(
+                    system_message="You are a helpful AI assistant.",
+                    user_text=selected_option,
+                    chat_history=st.session_state.chat_history
+                )
+                st.markdown(f"<div class='assistant-msg'><strong>HAL:</strong> {response}</div>", unsafe_allow_html=True)
 
-# Display follow-up question inside chat if available
-if st.session_state.response_ready and st.session_state.follow_up:
-    st.markdown(f"<div class='assistant-msg'><strong>HAL:</strong> {st.session_state.follow_up}</div>", unsafe_allow_html=True)
 
-    # Reset response state so user can type next input
-    st.session_state.response_ready = False