Toward an Open Source MLOps Architecture

A. Burgueño

doi:10.1109/MS.2024.3421675

2025-01-01

PRE

A. Burgueño *

Antonio Benítez‐Hidalgo

Cristóbal Barba-González

José F. Aldana‐Montes

DOI：10.1109/MS.2024.3421675

原文链接

原文求助

摘要

En 中文

We present a Kubernetes-based, open source MLOps framework to streamline the lifecycle management of machine learning models in production environments. We compared state-of-the-art MLOps tools and frameworks, demonstrating that ours meets the same features as proprietary options, such as Amazon SageMaker.

Keyword:

Pipelines

Computer architecture

Monitoring

Production

Testing

Python

Predictive models