Using Kubeflow Pipelines on Linode

Welcome! Here you will find a series of examples that demostrate setting up and using Kubeflow Pipelines on a Linode (now Akamai) cluster. The examples range from fairly simple to moderately complex scenarios that you can face in real-world settings. In these posts I will cover installation, debugging, handling input and output, conditional execution and branching. I hope you find this at least somewhat helpful!

Kubeflow Tutorials & Examples:

01. Markdown Visualization Pipeline

We will create a very simple pipeline to display some Markdown! This will help us later in visualizing and reporting metrics.

Go to code!

02. Simple MinIO Pipeline

We will connect to Kubeflow's built-in MinIO instance, download some data from internet, and store it in a Bucket.

Go to code!

03. Data Cleaning and Preparation Pipeline

Here we create a more practical data cleaning and preparation pipeline that processes a large CSV file containing car sale & advertisement data.

Go to code!

04. Model Training and Evaluation Pipeline

We use the data prepared in part 3 and train a baseline model as well three more practical models and compare their performance.

Go to code!