MLOps (Machine Learning Operations) Fundamentals
This course introduces participants to MLOps tools and best practices for deploying, evaluating, monitoring and operating production ML systems on Google Cloud.
What you'll learn
This course introduces participants to MLOps tools and best practices for deploying, evaluating, monitoring and operating production ML systems on Google Cloud. MLOps is a discipline focused on the deployment, testing, monitoring, and automation of ML systems in production. Machine Learning Engineering professionals use tools for continuous improvement and evaluation of deployed models. They work with (or can be) Data Scientists, who develop models, to enable velocity and rigor in deploying the best performing models.
Table of contents
- Introduction 1m
- Introduction to Containers 6m
- Containers and Container Images 7m
- Lab Intro 1m
- Pluralsight: Getting Started with GCP and Qwiklabs 4m
- Lab: Working with Cloud Build 0m
- Lab solution 5m
- Introduction to Kubernetes 4m
- Introduction to Google Kubernetes Engine 3m
- Compute Options Detail 10m
- Kubernetes Concepts 4m
- The Kubernetes Control Plane 6m
- Google Kubernetes Engine Concepts 6m
- Lab Intro 1m
- Lab: Deploying Google Kubernetes Engine 0m
- Lab solution 5m
- Deployments 4m
- Ways to Create Deployments 3m
- Services and Scaling 2m
- Updating Deployments 2m
- Rolling Updates 3m
- Blue-Green Deployments 5m
- Canary Deployments 8m
- Managing Deployments 1m
- Lab Intro 1m
- Lab: Creating Google Kubernetes Engine Deployments 0m
- Jobs and CronJobs 5m
- Parallel Jobs 4m
- CronJobs 4m
- System and concept overview 8m
- Describing a Kubeflow Pipeline with KF DSL 6m
- Pre-built components 5m
- Lightweight Python Components 3m
- Custom components 5m
- Compile, Upload and Run 5m
- Lab Intro: Continuous Training Pipeline with Kubeflow Pipeline and Cloud AI Platform 1m
- Lab: Continuous Training Pipeline with Kubeflow Pipeline and Cloud AI Platform 0m
- Lab Solution 39m