Explore ML Ops design patterns with Kubeflow Pipelines in this 20-minute conference talk by Amy Unruh from Google. Learn about the challenges of transitioning ML workflows from notebook exploration to production and discover effective patterns to address these issues. Gain insights into how Kubeflow Pipelines (KFP) can support and implement these patterns, with live demonstrations showcasing KFP in action. The talk covers an introduction to ML Ops challenges, an overview of Kubeflow pipelines, and practical pipeline examples to illustrate key concepts.
Overview
Syllabus
Intro
ML Ops challenges
Kubeflow pipelines
Pipeline examples
Second example
Taught by
CNCF [Cloud Native Computing Foundation]
Related Courses
-
ML Pipelines on Google Cloud
-
Build a Reproducible ML Workflow with Kubeflow Pipelines
-
A Deep Dive into Kubeflow Pipelines
-
Efficient Model Exploring and Continuous Delivery With Polyaxon + Kubeflow
-
Democratization of ML Pipelines and Bringing ML Workflows to Heterogeneous Cloud Native Platforms
-
MLOps Platforms: Amazon SageMaker and Azure ML
