Get in Touch
 Duration 28 hours

Course Outline

Introduction

Overview of Kubeflow Features and Components

  • Containers, manifests, and related elements.

Understanding Machine Learning Pipelines

  • Phases including training, testing, tuning, and deployment.

Installing Kubeflow on a Kubernetes Cluster

  • Preparing the execution environment (such as training and production clusters).
  • Acquiring, installing, and customising the system.

Executing Machine Learning Pipelines on Kubernetes

  • Creating a TensorFlow pipeline.
  • Creating a PyTorch pipeline.

Visualising Outcomes

  • Exporting and presenting pipeline performance metrics.

Adapting the Execution Environment

  • Adjusting the stack for various infrastructure needs.
  • Upgrading existing Kubeflow installations.

Utilising Kubeflow on Public Clouds

  • Integration with AWS, Microsoft Azure, and Google Cloud Platform.

Oversight of Production Workflows

  • Applying GitOps methodology in operations.
  • Organising and scheduling jobs.
  • Instantiating and managing Jupyter notebooks.

Requirements

  • Basic understanding of Python syntax
  • Practical experience with TensorFlow, PyTorch, or similar machine learning frameworks
  • An account with a public cloud provider (optional)

Target Audience

  • Developers
  • Data scientists

Related Categories