I want to do automatic labeling by adding an ML backend to Label Studio with my YOLO model. I created a Flask project. I added my model from Add model and it makes the predictions correctly. So, my Flask application works correctly. However, I can’t do automatic labeling in Label Studio. I send the predicted data to Label Studio in JSON format, but I can’t do automatic labeling. Can you please help? @sajarin
Thanks for posting here @Emre06,
Can you share the JSON of the predictions that are generated by your model?
You’ve already confirmed this elsewhere but for others reading this thread: Make sure that the model is correctly connected in Label Studio.
- Go to Settings > Machine Learning and verify that the model URL is correct and the connection is active.
- Also make sure the Interactive preannotations toggle is enabled.
What happens when you follow the instructions about calling your predict endpoint in this doc? Label Studio Documentation — Integrate Label Studio into your machine learning pipeline