Training-serving skew

As explained in Machine Learning Platform Engineering

Training-serving skew is a discrepancy between the feature transformations or data seen during model training and those applied during inference.

Training-serving skew occurs when feature transformations or other data seen during training differ from those applied during inference. The mismatch is easy to overlook and can undermine production predictions. Feature stores address it by serving consistently transformed features through offline storage for training or batch inference and online storage for real-time inference. section 1.3.3

Figure 1.7 Feature stores take in transformed data (features) as input and have facilities to store, catalog, and serve features.

Use the feature-store architecture to trace how transformed data becomes cataloged, stored, and served across training and inference paths. section 1.3.3

How Machine Learning Platform Engineering covers it

How deep it goes

Defines the discrepancy, identifies differing transformations as a cause, and stops short of detailed detection or remediation procedures.

Examples

Illustrates skew through different transformations during training and inference, including complex joins across multiple data sources.

What you can do after

Apply it via feature stores that serve transformed features through offline and online paths.

How it connects

Connects training-serving skew to feature stores, feature transformations, model training, batch inference, and real-time inference.

liveBookmanning.com

Read Machine Learning Platform Engineering on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Machine Learning System DesignA condition in which a machine learning model performs poorly because its training data does not represent the data encountered during application or serving.

Compare how 2 books explain training-serving skew →

Related concepts

• feature storeaddressed by · A feature store is a system for managing and serving machine-learning features for use in model training and inference.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning Platform Engineering. The section links point to the parts of the book it draws on.