Update app.py

app.py

@@ -1,32 +1,11 @@
-"""
 # Inference
 
-import gradio as gr
-
-app = gr.load(
-    "google/gemma-2-2b-it",
-    src = "models",
-    inputs = [gr.Textbox(label = "Input")],
-    outputs = [gr.Textbox(label = "Output")],
-    title = "Google Gemma",
-    description = "Inference",
-    examples = [
-        ["Hello, World."]
-    ]
-).launch()
-"""
-
-# Inference Client
-
 import gradio as gr
 from huggingface_hub import InferenceClient
-#import os
-
-#hf_token = os.getenv("HF_TOKEN")
 
 client = InferenceClient("google/gemma-2-2b-it")
 
-def fn_chat(
+def fn(
     message,
     history: list[tuple[str, str]],
     #system_message,
@@ -41,7 +20,8 @@ def fn_chat(
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
         if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+            #messages.append({"role": "assistant", "content": val[1]})
+            messages.append({"role": "bot", "content": val[1]})
 
     messages.append({"role": "user", "content": message})
 
@@ -60,7 +40,7 @@ def fn_chat(
         yield response
 
 app = gr.ChatInterface(
-    fn = fn_chat,
+    fn = fn,
     additional_inputs = [
         #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"),
@@ -68,7 +48,7 @@ app = gr.ChatInterface(
         gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
     ],
     title = "Google Gemma",
-    description = "Chatbot",
+    description = "Inference",
     examples = [
         ["Hello, World."]
     ]