DAG Card is the new Model Card

TL;DR

This paper introduces DAG Cards as a new documentation framework emphasizing data-centric AI and pipeline-level documentation, providing an open implementation to enhance real-world ML deployment transparency.

Contribution

It proposes DAG Cards as a novel documentation approach focused on ML pipelines, extending the concept of Model Cards for data-centric AI practices.

Findings

DAG Cards effectively document ML pipelines for deployment.

Open implementation facilitates adoption and reproducibility.

Highlights importance of pre- and post-training documentation.

Abstract

With the progressive commoditization of modeling capabilities, data-centric AI recognizes that what happens before and after training becomes crucial for real-world deployments. Following the intuition behind Model Cards, we propose DAG Cards as a form of documentation encompassing the tenets of a data-centric point of view. We argue that Machine Learning pipelines (rather than models) are the most appropriate level of documentation for many practical use cases, and we share with the community an open implementation to generate cards from code.