Get in Touch
 Duration 14 hours

Course Outline

Getting Started with Kubeflow

  • Understanding the mission and architectural design of Kubeflow
  • Overview of core modules and the broader ecosystem
  • Deployment strategies and platform features

Interacting with the Kubeflow Dashboard

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

Essentials of Kubeflow Pipelines

  • Pipeline architecture and component design
  • Creating pipelines using the Python SDK
  • Running, scheduling, and monitoring pipeline executions

Training ML Models on Kubeflow

  • Distributed training methodologies
  • Leveraging TFJob, PyTorchJob, and other operators
  • Resource allocation and autoscaling within Kubernetes

Serving Models via Kubeflow

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

Overseeing ML Workflows on Kubernetes

  • Managing versions for data, models, and artefacts
  • Incorporating CI/CD into ML pipelines
  • Security and role-based access control

Best Practices for Production-Grade ML

  • Designing resilient workflow structures
  • Ensuring observability and monitoring
  • Resolving common Kubeflow challenges

Advanced Concepts (Optional)

  • Multi-tenant Kubeflow configurations
  • Hybrid and multi-cluster deployment strategies
  • Extending Kubeflow with custom components

Wrap-Up and Future Directions

Requirements

  • A grasp of containerised applications
  • Proficiency in basic command-line operations
  • Knowledge of fundamental Kubernetes concepts

Target Audience

  • Machine learning practitioners
  • Data scientists
  • DevOps teams new to the Kubeflow platform

Number of participants


Price per participant

Testimonials (4)

Provisional Upcoming Courses (Require 5+ participants)

Related Categories