Monitoring Models and Infrastructure
About the course
Monitoring your models and infrastructure is important to ensure the quality of predictive output of your models. This two-day module will teach you the three different types of monitoring that will ensure you to keep AI running at scale with high performance. The three types of monitoring are performance monitoring, data drift monitoring and infrastructure monitoring.
Performance monitoring allows you to deep dive into the performance of the models currently running on production. We will explore how to monitor this performance in model control centers and how you can build them. While performance monitoring is all about looking back, we will also teach you how to proactively identify models that might start to lose their predictive power. We will do this through monitoring data drift, which will allow you to proactively signal which models might be giving less trustworthy predictions. Additionally, it is also important to ensure that the infrastructure on which your models are running is performant, for which we will teach you how to monitor your infrastructure. Finally, we will investigate tools like Promotheus and Grafana to bring everything together
Why this is for you
Have you been confronted with models you operationalized slowly losing their predictive power over time? Do you feel overwhelmed by the increasing number of models running in production? Would you like to proactively signal models that are vulnerable to lose their predictive power? This course will teach you how to monitor models and the underlying infrastructure so you can ensure they are continuously providing high performances over time.
Theory and practical use
All trainings in the GAIn portfolio combine high-quality standardized training material with theory sessions from experts and hands-on experience where you directly apply the material to real-life cases. Each training is developed by top of the field practitioners which means they are full of industry examples along with practical challenges and know-how, fueling the interactive discussions during training. We believe this multi-level approach creates the ideal learning environment for participants to thrive.