Update app.py

app.py

@@ -29,25 +29,35 @@ print("Loading the model...")
 llm = Llama(model_path=model_path)
 print("Model loaded successfully!")
 
-def generate_response(message, history, temperature=0.7, top_p=1.0, max_tokens=256):
-    # history_langchain_format = []
-    # for human, ai in history:
-    #     history_langchain_format.append(HumanMessage(content=human))
-    #     history_langchain_format.append(AIMessage(content=ai))
-
-    # history_langchain_format.append(HumanMessage(content=message))
-    
+def generate_response(message, history, temperature=0.7, top_p=1.0, max_tokens=256):  
+     """
+    Generate a response based on the user's message and the conversation history.
+    """
     try:
-        gpt_response = llm(message, max_tokens=max_tokens, temperature=temperature, top_p=top_p)
-        return gpt_response["choices"][0]["text"].strip()
+        conversation = ""
+        for msg in history:
+            role, content = msg
+            conversation += f"<start_of_turn>{role}\n{content}\n<end_of_turn>"
+
+        conversation += f"<start_of_turn>user\n{message}\n<end_of_turn>"
+
+        response = llm(conversation, max_tokens=max_tokens, temperature=temperature, top_p=top_p)
+        return response["choices"][0]["text"].strip()
+
     except Exception as e:
-        return f"Error generating response: {e}"
+        return f"Error generating response: {str(e)}"
 
 if __name__ == "__main__":
-    rag = gr.ChatInterface(
+    gguf_demo = gr.ChatInterface(
         generate_response,
         title=title,
         description=description,
     )
 
-    rag.launch(share=True)
+    gguf_demo.launch(share=True)
+
+    try:
+        if llm:
+            llm.close()
+    except Exception as e:
+        print(f"Error closing model: {e}")