Update app.py

app.py

@@ -5,6 +5,9 @@ import requests
 import time
 import os
 
+def start_server():   
+  os.system("uvicorn InferenceServer:app --port 8080 --host 0.0.0.0 --workers 2")
+
 def load_models():
   if not is_port_in_use(8080):
     with st.spinner(text="The model is loading."):