Update app.py

app.py

@@ -1,33 +1,46 @@
 import os
-import re
 from dotenv import load_dotenv
+import gradio as gr
 from huggingface_hub import InferenceClient
+
 load_dotenv()
-HF_Token=os.getenv("HUGGINGFACEHUB_ACCESS_TOKEN")
-
-Model_id="google/gemma-2-2b-it"
-client=InferenceClient(model=Model_id, token=HF_Token)
-message=[{"role": "system", "content": "Hello, my name is John. What is your name?"}]
-
-print("type exit to quit.\n")
-
-while True:
-     try:
-    user_text=input("you: ").strip()
-     except EOFError:
-        print("\nInput stream closed. Exiting...")
-        break
-
-    if(user_text.lower()=="exit"):
-        print("good bye")
-        break
-    message.append({"role": "user", "content": user_text})
-   
-        resp=client.chat_completion(messages=message,max_tokens=100,temperature=.7,stop=["\nUser","\nSystem: "])
-        bot_text=resp.choices[0].message["content"]
+
+HF_Token = os.getenv("HUGGINGFACEHUB_ACCESS_TOKEN")
+MODEL_ID = "google/gemma-2-2b-it"
+
+client = InferenceClient(
+    model=MODEL_ID,
+    token=HF_Token
+)
+
+messages = [
+    {"role": "system", "content": "Hello, my name is John. What is your name?"}
+]
+
+def chat(user_text):
+    if not user_text:
+        return ""
+
+    messages.append({"role": "user", "content": user_text})
+
+    try:
+        resp = client.chat_completion(
+            messages=messages,
+            max_tokens=150,
+            temperature=0.7
+        )
+        bot_text = resp.choices[0].message["content"]
+        messages.append({"role": "assistant", "content": bot_text})
+        return bot_text
     except Exception as e:
-        print("bot Error:", e)
-        message.pop()
-        continue
-    print("bot:", bot_text+"\n")
-    message.append({"role": "assistant", "content": bot_text})
+        messages.pop()
+        return f"Error: {e}"
+
+demo = gr.Interface(
+    fn=chat,
+    inputs=gr.Textbox(label="You"),
+    outputs=gr.Textbox(label="Bot"),
+    title="Gemma Chatbot (Hugging Face Space)"
+)
+
+demo.launch()