Monitor Model Drift

See Extended Examples for complete configuration files and templates.

Detect and alert on data drift and concept drift in production ML models using statistical tests and automated monitoring.

When to Use

• Production ML models experiencing unexplained performance degradation
• New data distributions differ from training data
• Seasonal or temporal shifts in input features
• Need proactive alerts before business metrics are impacted
• Regulatory requirements for model monitoring (e.g., SR 11-7, EU AI Act)
• Multiple model versions deployed requiring drift comparison

Inputs