From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

ML Lens for monitoring

ML Lens for monitoring

From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep

Start my 1-month free trial Buy for my team

ML Lens for monitoring

“

(soft music) - [Instructor] Hello guys, and welcome again. In today's lesson, we're going to talk about the AWS machine learning lens for monitoring. So the monitoring phase in the machine learning life cycle ensures that a deployed model continues to perform as expected over time. Think of it as a feedback loop, maintaining the model's alignment with the real world conditions. The key goals of the monitoring phase would be to track the data drift and the model drift. The data drift is the changes in the input data distribution compared to the data used during training, and the model drift identifies the degradation in the model performance due to changes in the underlying patterns. The second key goal is to alert us on anomalies. So it'll detect issues such as data quality problems or bias or explainability, and to also trigger re-training pipelines, so to automatically initiate re-training workflows when drift or anomalies are detected. So the alignment with the AWS Well architected…

Contents