Data drift occurs all the time. But when does it impact the quality of your ML predictions? In this talk, you’ll learn how to detect data drift and precisely quantify its impact on model performance. In the first part, we will cover the two leading causes of model failure and discuss how data drift can impact performance over time. In the second part, we will discuss novel open-source algorithms that allow us to quantify how data drift impacts performance for regression and classification tasks.
March 26, 2024 4:00 PM
Wojtek Kuberski