Chapter 12: Reducing Time to Insights for Anomaly Detections
In the previous chapters, you learned how to prepare multivariate datasets, how to train and evaluate an anomaly detection model, and how to configure an inference scheduler. To get the most from Amazon Lookout for Equipment, you can partner with a data engineer or a data scientist who will help you improve your model performance and go further in the post-processing of results.
The main objectives of this chapter are to point you in the right direction to visualize and monitor your models. This will be very valuable to detect any drift that would trigger either retraining or further investigation. In addition, you will learn how to build an automation pipeline, which will be critical to iterate as fast as possible without having to manually navigate through multiple console screens.
In this chapter, we're going to cover the following main topics:
- Improving your model's accuracy
- Processing the...