ParentFull threadandbberger·if you're doing vision based obstacle avoidance and you're training your model on specific classes of objects and not just object you're doing it wrongView on HN