Get in Touch
 Duration 14 hours

Course Outline

Introduction to Kubeflow

  • Grasping the Kubeflow mission and underlying architecture
  • Overview of core components and the broader ecosystem
  • Exploring deployment options and platform features

Utilizing the Kubeflow Dashboard

  • Navigating the user interface
  • Managing notebooks and workspaces
  • Connecting storage and data sources

Kubeflow Pipelines Basics

  • Structuring pipelines and designing components
  • Creating pipelines using the Python SDK
  • Running, scheduling, and tracking pipeline executions

Training ML Models on Kubeflow

  • Strategies for distributed training
  • Utilizing TFJob, PyTorchJob, and other operators
  • Handling resource management and autoscaling in Kubernetes

Model Serving with Kubeflow

  • Introduction to KFServing / KServe
  • Deploying models with custom runtimes
  • Handling revisions, scaling, and traffic routing

Orchestrating ML Workflows on Kubernetes

  • Versioning data, models, and artifacts
  • Integrating CI/CD for ML pipelines
  • Implementing security and role-based access control

Best Practices for Production ML

  • Designing dependable workflow patterns
  • Focusing on observability and monitoring
  • Resolving common Kubeflow challenges

Advanced Topics (Optional)

  • Setting up multi-tenant Kubeflow environments
  • Managing hybrid and multi-cluster deployment scenarios
  • Expanding Kubeflow with custom components

Recap and Future Directions

Requirements

  • A foundational understanding of containerized applications
  • Proficiency with basic command-line operations
  • Familiarity with fundamental Kubernetes concepts

Target Audience

  • ML practitioners
  • Data scientists
  • DevOps teams looking to integrate Kubeflow

Testimonials (4)

Related Categories