liveBook · Manning

Contract

As explained in Delivering Machine Learning Projects: From design to deployment

An API is a defined interface that enables software components to communicate and invoke available data or functionality.

In production data infrastructure, an API can provide an access layer that invokes queries without exposing the underlying databases. This decouples the system from source implementation details and makes database upgrades easier to manage. section 9.4.1

When building a production data pipeline:

  • Route calls and interactions through a well-controlled, documented API, especially when the existing architecture exposes databases directly. section 9.4.1
  • Test the resulting data pipeline: changes in production can introduce biases that affect model behavior. section 9.4.1

How Delivering Machine Learning Projects: From design to deployment covers it

How deep it goes
Explains the abstraction’s benefits and recommends a controlled, documented API, but does not detail its implementation.
Examples
Describes an API address that invokes a query and a database upgrade managed through API calls.
What you can do after
Evaluate how an API layer can isolate production systems from database implementation changes.
How it connects
Connects APIs with databases, data pipelines, and machine learning model fidelity.

Also explained in

Compare how 44 books explain contract →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Delivering Machine Learning Projects: From design to deployment. The section links point to the parts of the book it draws on.