Update app.py

app.py

@@ -2,10 +2,10 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
-# ✅ Load token from secrets (never hardcode!)
-HF_TOKEN = os.environ.get('telemedpro')
+# 🔹 Hardcoded token for internal use (or use os.environ for safety)
+HF_TOKEN = os.getenv('telemedpro')
 
-# Persona system prompt
+# Persona system message
 PERSONA_MESSAGE = (
     "You are Dr. Alex, a highly knowledgeable yet empathetic doctor. "
     "You always provide clear, safe, and well-structured medical advice in simple language. "
@@ -13,39 +13,47 @@ PERSONA_MESSAGE = (
     "You behave politely, patiently, and with care, like a trusted family doctor."
 )
 
-# Create HF InferenceClient (safe)
+# Create HF InferenceClient
 client = InferenceClient(
     model="m42-health/Llama3-Med42-70B",
     token=HF_TOKEN
 )
 
+# Respond function with safe error handling
 def respond(message, history, max_tokens, temperature, top_p):
-    # Always include persona system prompt
-    messages = [{"role": "system", "content": PERSONA_MESSAGE}]
+    try:
+        # Initialize messages with persona
+        messages = [{"role": "system", "content": PERSONA_MESSAGE}]
 
-    # Add conversation history
-    for h in history:
-        messages.append({"role": "user", "content": h[0]})
-        messages.append({"role": "assistant", "content": h[1]})
+        # Add conversation history safely
+        for h in history:
+            user_msg = h[0] if h[0] is not None else ""
+            ai_msg = h[1] if h[1] is not None else ""
+            messages.append({"role": "user", "content": user_msg})
+            messages.append({"role": "assistant", "content": ai_msg})
 
-    # Add user message
-    messages.append({"role": "user", "content": message})
+        # Add current user message
+        messages.append({"role": "user", "content": message})
 
-    # Collect streamed response
-    response = ""
-    for msg in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        if msg.choices and msg.choices[0].delta.content:
-            token = msg.choices[0].delta.content
-            response += token
-            yield response
+        # Stream AI response token by token
+        response = ""
+        for msg in client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p
+        ):
+            if msg.choices and msg.choices[0].delta.content:
+                token = msg.choices[0].delta.content
+                response += token
+                yield response
 
-# 🔹 Chat Interface
+    except Exception as e:
+        # Return error as a string so Space never crashes
+        yield f"⚠️ Space error: {e}"
+
+# 🔹 Gradio Chat Interface
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
@@ -53,13 +61,15 @@ chatbot = gr.ChatInterface(
         gr.Slider(1, 2048, value=512, step=1, label="Max tokens"),
         gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
-    ],
+    ]
 )
 
+# 🔹 Layout
 with gr.Blocks() as demo:
     with gr.Row():
         gr.Markdown("## 🩺 AI Health Mentor — Dr. Alex")
     chatbot.render()
 
+# 🔹 Launch the Space safely
 if __name__ == "__main__":
     demo.launch(show_error=True)