Introduction¶
Learn how to use the model to label new data using Label Studio and retrain the model iteratively.
Requirements¶
The following requirements are necessary to follow this part in addition to those described in the first part:
- A Chrome or a Firefox based browser for better compatibility
State of the MLOps process¶
Production feedback and new data require a way to improve the model iteratively. In this part, you will address the following issues:
- Labeling of supplemental data is not systematic or uniform
- Labeling of supplemental data is time intensive
- Model needs to be retrained using higher-quality data
Note
This part focuses on labeling data locally so you can experiment without extra
infrastructure. For the same reason, we will run dvc repro on your local
machine.
In a production setup, you should push the new labeled data to a branch and let your CI/CD pipeline retrain the model on the Kubernetes cluster, as set up in Part 3 - Serve and deploy.