← MLOps basics
Lesson 2 of 3
Monitoring
Watch, read the card, then practise. Data drift · Model monitoring
The video for this lesson is being recorded.
Meanwhile, the notes below and the 60-second cards cover the same ideas.
Practise this topicData drift
Data drift is when live inputs start to look different from the training data, so the model's accuracy can quietly drop.
Remember: Models age. Plan to monitor and retrain.
Model monitoring
Model monitoring watches both the system (speed, errors) and the model (inputs, outputs, accuracy) in production.
Remember: A sudden change in predictions is a warning, not good news.
Try the example in Code Lab
import numpy as np train, live = np.array([30, 32, 28]), np.array([48, 52, 50]) print(round(float(live.mean() - train.mean()), 1), bool(abs(live.mean() - train.mean()) > 2 * train.std()))