Overview of the ML workflow

Kubeflow aims to be your Kubernetes ML toolkit. The ML tools that are required for your workflow can then be specified using the Kubeflow configurations and the workflow can be deployed to various platforms for testing and production use as required.

Let’s have a look at the Kubeflow components before we get into the intricacies of ML workflows.

Introduction – Kubeflow and its components

Kubeflow is a system for deploying, scaling, and managing complex systems based on Kubernetes. For data scientists, Kubeflow is the go-to platform for building and testing ML pipelines. It is also for ML developers and operations teams who wish to deploy ML systems in a variety of contexts for development, testing, and production.

Kubeflow is a framework for establishing the components of your ML system on top of Kubernetes, as shown in the following diagram:

Figure 9.1 – Kubeflow components on top of Kubernetes

...