Evaluation

What is a feature store?

A feature store serves consistent numeric or categorical features to training and to inference, usually for tabular models.

Updated

How it works

  • Features are defined once.
  • Training and serving read the same definitions.
  • The values are typically scalars, not frames.

What it is not

It is not computed columns on images and video, and it is not an embedding index.

feature store: this, and the thing it is confused with

feature store: this, and the thing it is confused with
ThisNot this
ValuesScalars and categoriesFrames and documents
PointTrain/serve consistencyMedia pipelines
OverlapBoth cache derived valuesThey are the same product

Where Pixeltable fits

Pixeltable computed columns can feed features, including from media. A feature store such as Feast or Tecton is the tabular product. They answer different queries.

Questions

How does feature store work?
Features are defined once. Training and serving read the same definitions. The values are typically scalars, not frames.
What is feature store often confused with?
It is not computed columns on images and video, and it is not an embedding index.