Change is the only constant in life - Heraclitus quote is so relevant in the Machine Learning world. The ML models that are running in the production need to be monitored. Why? Because data can change in time, leading to wrong predictions. The change in data distribution is so-called a data drift. I will show you how to build a dashboard for data drift detection in Python. I will use Evidently Python package for detecting data drift and building a dashboard in the Jupyter Notebook. The dashboard notebook will be published as a web application with the Mercury framework.

Read More