I am trying to create an object detection framework which takes input from user and creates custom object detection models based on the user selection. For this I have to dy