What is Model Drift?
MLOps
Model Drift — When a deployed model performance degrades over time because the real-world data distribution changes from what the model was trained on. Requires monitoring and retraining.
FAQ
What is model drift?
When your model gets worse over time because real-world data changes. Example: a fraud model trained on 2024 patterns failing on 2026 fraud tactics.
How to detect drift?
Monitor prediction distributions, accuracy metrics, and input feature distributions. Tools: Evidently AI, WhyLabs, custom dashboards.
Related Terms
Learn Model Drift in depth
Free hands-on course with code examples and Google Colab notebooks.
Start Course →