Update app.py

app.py

@@ -3,6 +3,7 @@ import time
 import requests
 from streamlit.components.v1 import html
 import os
+import ollama
 
 # Import transformers and cache the help agent for performance
 @st.cache_resource
@@ -234,41 +235,39 @@ def ask_llama(conversation_history, category, is_final_guess=False):
 # New function for the help AI assistant using the Hugging Face InferenceClient
 def ask_help_agent(query):
     try:
-        from huggingface_hub import InferenceClient
-        # Initialize the client with the provided model
-        client = InferenceClient("HuggingFaceH4/zephyr-7b-beta", token=os.environ.get("HF_HUB_TOKEN"))
+        import ollama
+        
+        # System message setup
         system_message = "You are a friendly Chatbot."
-
-        # Build history from session state (if any)
-        history = []
+        
+        # Build conversation history in Ollama format
+        messages = [{"role": "system", "content": system_message}]
+        
         if "help_conversation" in st.session_state:
             for msg in st.session_state.help_conversation:
-                # Each history entry is a tuple: (user query, assistant response)
-                history.append((msg.get("query", ""), msg.get("response", "")))
-
-        messages = [{"role": "system", "content": system_message}]
-        for user_msg, bot_msg in history:
-            if user_msg:
-                messages.append({"role": "user", "content": user_msg})
-            if bot_msg:
-                messages.append({"role": "assistant", "content": bot_msg})
+                if msg.get("query"):
+                    messages.append({"role": "user", "content": msg["query"]})
+                if msg.get("response"):
+                    messages.append({"role": "assistant", "content": msg["response"]})
+        
+        # Add current query
         messages.append({"role": "user", "content": query})
-
-        response_text = ""
-        # Using streaming to collect the entire response from the model
-        for message in client.chat_completion(
-            messages,
-            max_tokens=150,
-            stream=True,
-            temperature=0.7,
-            top_p=0.95,
-        ):
-            token = message.choices[0].delta.content
-            response_text += token
-        return response_text
+        
+        # Get response from Ollama
+        response = ollama.chat(
+            model="llama3",  # Can also use "mistral", "phi3", etc.
+            messages=messages,
+            options={
+                "temperature": 0.7,
+                "top_p": 0.95,
+                "num_predict": 150  # Similar to max_tokens
+            }
+        )
+        
+        return response['message']['content']
+        
     except Exception as e:
         return f"Error in help agent: {str(e)}"
-
 # Main game logic with enhanced UI
 def main():
     inject_custom_css()