Feature Stores: A Key Enabler for Feature Reusability and Availability Across Machine Learning Pipelines

Feature Stores: A Key Enabler for Feature Reusability and Availability Across Machine Learning Pipelines

This article presents a technical analysis of the feature store (FS) from the perspective of machine learning (ML) model development and highlights its potential benefits. We propose key considerations in developing a workable and efficient FS for ML pipelines.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computer (Long Beach, Calif.) Calif.), 2024-01, Vol.57 (1), p.69-74
Hauptverfasser: Majeed, Abdul, Hwang, Seong Oun, Lin, Hsiao-Ying
Format: Artikel
Sprache:eng
Schlagworte:
Machine learning
Pipelines
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This article presents a technical analysis of the feature store (FS) from the perspective of machine learning (ML) model development and highlights its potential benefits. We propose key considerations in developing a workable and efficient FS for ML pipelines.
ISSN:0018-9162
1558-0814
DOI:10.1109/MC.2023.3308868