make start faster

Azaion.Inference/main.pyx

@@ -32,10 +32,10 @@ cdef class CommandProcessor:
         self.inference_queue = Queue(maxsize=constants.QUEUE_MAXSIZE)
         self.remote_handler.start()
         self.running = True
-        model = self.api_client.load_ai_model()
-        self.inference = Inference(model, self.on_annotation)
+        self.inference = Inference(self.api_client, self.on_annotation)
 
     def start(self):
+        print('Started!')
         while self.running:
             try:
                 command = self.inference_queue.get(timeout=0.5)