Update app.py

app.py

@@ -6,12 +6,11 @@ from langchain_core.prompts import PromptTemplate
 from langchain_core.output_parsers import StrOutputParser
 from transformers import pipeline
 
-# Use the HF_TOKEN environment variable (make sure it's set in your Hugging Face Space)
+# Use environment variables for keys
 HF_TOKEN = os.getenv("HF_TOKEN")
 if HF_TOKEN is None:
     raise ValueError("HF_TOKEN environment variable not set. Please set it in your Hugging Face Space settings.")
 
-# Use environment variable for the NASA API key too
 NASA_API_KEY = os.getenv("NASA_API_KEY")
 if NASA_API_KEY is None:
     raise ValueError("NASA_API_KEY environment variable not set. Please set it in your Hugging Face Space settings.")
@@ -19,8 +18,9 @@ if NASA_API_KEY is None:
 # Set up Streamlit UI
 st.set_page_config(page_title="HAL - NASA ChatBot", page_icon="🚀")
 
-# --- Ensure Session State Variables are Initialized ---
+# --- Initialize Session State Variables ---
 if "chat_history" not in st.session_state:
+    # Initial greeting stored in chat history
     st.session_state.chat_history = [{"role": "assistant", "content": "Hello! How can I assist you today?"}]
 
 if "response_ready" not in st.session_state:
@@ -30,7 +30,6 @@ if "follow_up" not in st.session_state:
     st.session_state.follow_up = ""  # Stores follow-up question
 
 # --- Set Up Model & API Functions ---
-# Specify your intended model here. If you need to change it, update the model_id.
 model_id = "mistralai/Mistral-7B-Instruct-v0.3"
 
 # Initialize sentiment analysis pipeline with explicit model specification
@@ -41,7 +40,7 @@ sentiment_analyzer = pipeline(
 )
 
 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.
+    # Specify task="text-generation" so that the endpoint uses the right model function.
     return HuggingFaceEndpoint(
         repo_id=model_id,
         max_new_tokens=max_new_tokens,
@@ -82,7 +81,8 @@ def generate_follow_up(user_text):
 
 def get_response(system_message, chat_history, user_text, max_new_tokens=256):
     """
-    Generates HAL's response, making it more conversational and engaging.
+    Generates HAL's response in a friendly, conversational manner.
+    The prompt instructs the model to ignore previous greetings and focus on the new user question.
     """
     sentiment = analyze_sentiment(user_text)
     action = predict_action(user_text)
@@ -99,13 +99,14 @@ def get_response(system_message, chat_history, user_text, max_new_tokens=256):
 
     hf = get_llm_hf_inference(max_new_tokens=max_new_tokens, temperature=0.9)
 
+    # Updated prompt: Instruct the model not to repeat previous greetings.
     prompt = PromptTemplate.from_template(
         (
             "[INST] {system_message}\n\nCurrent Conversation:\n{chat_history}\n\n"
             "User: {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:"
+            "AI: Please answer the user's question without repeating any previous greetings. "
+            "Keep your response friendly and conversational, starting with a phrase like "
+            "'Certainly!', 'Of course!', or 'Great question!'.\nHAL:"
         )
     )
     chat = prompt | hf.bind(skip_prompt=True) | StrOutputParser(output_key='content')