Update app.py

app.py

@@ -1,30 +1,22 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-from unsloth.chat_templates import get_chat_template
 
 # Initialize the InferenceClient with the appropriate model
 client = InferenceClient("wop/kosmox")
 
-# Define the chat template and tokenizer configuration
-tokenizer = get_chat_template(
-    tokenizer=None,  # Assuming you need to pass an actual tokenizer here
-    chat_template="phi-3",
-    mapping={"role": "from", "content": "value", "user": "human", "assistant": "gpt"},
-)
-
 def format_messages(system_message, history, user_message):
     # Create a formatted string according to the specified chat template
     formatted_message = "<s>\n"
     if system_message:
-        formatted_message += f"{system_message}\n"
+        formatted_message += f"<|system|>\n{system_message}\n"
 
     for user_msg, assistant_msg in history:
         if user_msg:
-            formatted_message += f"{user_msg}\n"
+            formatted_message += f"<|user|>\n{user_msg}\n"
         if assistant_msg:
-            formatted_message += f"{assistant_msg}\n"
+            formatted_message += f"<|assistant|>\n{assistant_msg}\n"
     
-    formatted_message += f"{user_message}\n"
+    formatted_message += f"<|user|>\n{user_message}\n"
     return formatted_message
 
 def respond(
@@ -56,7 +48,7 @@ def respond(
 demo = gr.ChatInterface(
     fn=respond,
     additional_inputs=[
-        gr.Textbox(value="You are AI.", label="System message"),
+        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
@@ -64,4 +56,4 @@ demo = gr.ChatInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()