Concept drift is a change in the relationship between model inputs and the target behavior being predicted. In production, monitoring pipelines detect performance degradation and alert developers or trigger retraining. section 1.1.1
A production workflow handles drift by:
- Periodically assessing datasets, model performance, and model-serving behavior. section 1.1.1
- Alerting developers when model performance degrades. section 1.1.1
- Retraining the model when monitoring identifies a production problem. section 1.1.1
Figure 1.2 MLOps applies DevOps approaches to deep learning for the productionization phase, when models get shipped to production. (Source: Machine Learning Engineering in Action, by Ben Wilson, Manning, 2022, figure 2.7)
Use the MLOps architecture to place drift detection alongside continuous training, deployment, model serving, and monitoring. section 1.1.1
