Index Ai Academy
Glossary · Phase 11 · Professional practice

Feature store

Also written: feature stores

A system that defines each feature once and serves it both to training, as of any past time, and to the live model, as of now.

Try it now

Pick a case. Only the scaler's statistics differ between them.

Learn it properly

Feature pipelines and training-serving skew · Why a model that passes offline fails live: statistics refit at serving, two code paths, and features from the future, with the fix for each. About 40 minutes.

Other terms from the same lesson

  • parity test: A test that runs the same raw rows through the training path and the serving path and compares the features they produce.
  • point-in-time correctness: Building each training row's features as of the moment its prediction would have been made, using only earlier events.
  • training-serving skew: Any difference between how a feature is computed for training and how it is computed when the model serves requests. It gives wrong predictions without an error.
Keep learning

More in plain English.

1286 terms in the glossary, each with the lesson that teaches it.

1286 terms

The glossary

Every word you'll meet in AI and machine learning, explained in everyday language.

Phase 11 · lesson 40 min

Feature pipelines and training-serving skew

Why a model that passes offline fails live: statistics refit at serving, two code paths, and features from the future, with the fix for each.

Never trained an AI? Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account. Teach it to write