Update app.py

app.py

@@ -29,11 +29,13 @@ print("Loading the model...")
 llm = Llama(model_path=model_path)
 print("Model loaded successfully!")
 
+# Define the function to generate responses
 def generate_response(messages, temperature=0.7, top_p=1.0, max_tokens=256):
     """
     Generate a response from the model.
     Args:
-        messages (list): List of conversation history in a tuple format (user, assistant).
+        messages (list): List of conversation history in the format:
+                         [{'role': 'user', 'content': '...'}, {'role': 'assistant', 'content': '...'}]
         temperature (float): Sampling temperature.
         top_p (float): Top-p sampling parameter.
         max_tokens (int): Maximum number of tokens to generate.
@@ -41,11 +43,16 @@ def generate_response(messages, temperature=0.7, top_p=1.0, max_tokens=256):
         str: The model's response.
     """
     prompt = ""
-    for user_message, assistant_message in messages:
-        prompt += f"<start_of_turn>user\n{user_message}\n<end_of_turn>"
-        prompt += f"<start_of_turn>model\n{assistant_message}\n<end_of_turn>"
-    
-    prompt += "<start_of_turn>user\n" + messages[-1][0] + "\n<end_of_turn>"
+    for message in messages:
+        role = message['role']
+        content = message['content']
+        
+        if role == 'user':
+            prompt += f"<start_of_turn>user\n{content}\n<end_of_turn>"
+        elif role == 'assistant':
+            prompt += f"<start_of_turn>model\n{content}\n<end_of_turn>"
+
+    prompt += "<start_of_turn>user\n" + messages[-1]['content'] + "\n<end_of_turn>"
 
     try:
         response = llm(prompt, max_tokens=max_tokens, temperature=temperature, top_p=top_p)