EasyDeepLearn

What do you monitor about features (not just models)?

medium

Answer

  • (1) Freshness: is the feature updated on schedule?
  • (2) Completeness: missing rate per feature.
  • (3) Distribution drift: PSI / MMD vs baseline.
  • (4) Cardinality: for categorical, new unseen values.
  • (5) Correlation stability: pairwise correlations vs baseline.
  • (6) Feature importance stability: SHAP over time.
  • (7) Range violations: values outside expected.
  • Store per-feature health score → dashboard.
  • Predictive: feature health degrades before model does.
Check yourself — multiple choice
  • Random
  • Freshness + completeness + drift (PSI/MMD) + cardinality (new values) + correlation stability + SHAP importance stability + range violations; feature health degrades before model
  • Only models
  • Not real

Feature monitoring: freshness + completeness + drift + cardinality + correlation + importance + range.

#monitoring#features

Practise MLOps & Data Quality

215 interview questions in this topic.

Related questions