Charmed Kubeflow delivers a powerful, sophisticated end-to-end MLOps platform which you can deploy in half an hour or less, using MicroK8s or another conformant Kubernetes distribution.
Charmed Kubeflow offers a radically transformative approach to building data engineering pipelines and AI/ML workflows based on containers, Kubernetes, open source software and the internet. As such the Charmed Kubeflow platform can dramatically amplify productivity levels for data scientists and engineers working with advanced analytics and AI.
Kubeflow Pipelines is the powerful workflow engine at the heart of the Charmed Kubeflow MLOps platform. The pipelines are defined using the ubiquitous YAML markup language, with pipeline steps developed in a Python DSL. Whilst incredibly powerful and highly optimised for rigorous GitOps-driven environments, the learning curve can be deterring for some data scientists.
In this tutorial, we will examine Elyra, a visual workflow design engine for Kubeflow Pipelines that can help to get you started with Charmed Kubeflow quickly and easily.
What you’ll learn
- How to get up and running with Elyra
- Some of the basic features of Elyra
- How to run a workflow with Elyra
- How Elyra works with Kubeflow Pipelines
- How to build a simple Elyra workflow
What you’ll need
- A Kubernetes cluster with one or more nodes running Charmed Kubeflow - see the install guide to get up and running
- Some command line knowledge
- Some knowledge of data science and data engineering concepts and technologies