skip to main content

Introduction

Kubeflow is designed to package the many components used in ML applications into cloud-native components, integrate the components, and exploit all the portability and management features that Kubernetes brings to the compute environment.

Highlights

  • A Kubeflow package (container) comprises integrated ML components.

Features and Benefits

  • Learn about the lifecycle of machine learning applications.
  • Understand how using Kubernetes and containers can help improve the management of the ML application lifecycle.

Key questions answered

  • Why is the lifecycle for ML different from other software applications?
  • What is the role of connectors in the Kubeflow project?

Table of contents

Ovum view

  • Summary
  • Containerized solution for the ML application lifecycle
  • A Kubeflow package (container) comprises integrated ML components

Appendix

  • Author