Deploying Kubeflow Pipelines with the Kale SDK

Dive In
Using the Kale SDK

Start Course


This course focuses on using the Kale SDK to decorate Python code snippets and generate and deploy Kubeflow Pipelines. Throughout this course you will explore, via hands on tutorials and challenges, how to convert Python files into Kubeflow Pipelines. SDK Concepts will be presented and demonstrated in a series of increasingly more advanced sections. By the end of this course, you will be fully prepared to begin deploying your own Kubeflow Pipelines based on Python code.

Kubeflow as a Service Usage

Because it’s the simplest way to get started, we will use Kubeflow as a Service as our Kubeflow environment. We’ll teach you how to organize and annotate cells in a Jupyter Notebook to define a Kubeflow Pipeline that will run on a Kubernetes cluster. This does not require any specialized knowledge of Kubernetes. Instead, we’ll use the open-source Kale JupyterLab extension.


We strongly recommend completing the Getting Started with Kubeflow and Deploying Kubeflow Pipelines with the Kale UI courses in advance of taking this course if you are not yet familiar with Kubeflow. Additionally we expect that you are familiar with Python and popular data science Python libraries.