MLOps With Red Hat OpenShift: A Cloud-Native Approach to Machine Learning Operations

Cover
Title Page
Copyright and Credits
Contributors
Table of Contents
Preface
Part 1: Introduction
Chapter 1: Introduction to MLOps and OpenShift
	What is MLOps?
	Introduction to OpenShift
		OpenShift features
	Understanding operators
	Understanding how OpenShift supports MLOps
	Red Hat OpenShift Data Science (RHODS)
	The advantages of the cloud
	ROSA
	Summary
	References
Part 2: Provisioning and Configuration
Chapter 2: Provisioning an MLOps Platform in the Cloud
	Technical requirements
	Installing OpenShift on AWS
		Preparing AWS accounts and service quotas
		Preparing AWS for ROSA provisioning
		Installing ROSA
		Adding a new machine pool to the cluster
	Installing Red Hat ODS
	Installing partner software on RedHat ODS
	Installing Pachyderm
	Summary
Chapter 3: Building Machine LearningModels with OpenShift
	Technical requirements
	Using Jupyter Notebooks in OpenShift
	Provisioning an S3 store
	Using ML frameworks in OpenShift
	Using GPU acceleration for model training
	Enabling GPU support
	Building custom notebooks
		Creating a custom notebook image
		Importing notebook images
	Summary
Part 3: Operating ML Workloads
Chapter 4: Managing a Model Training Workflow
	Technical requirements
	Configuring Pachyderm
	Versioning your data with Pachyderm
	Training a model using Red Hat ODS
	Building a model training pipeline
		Installing Red Hat OpenShift Pipelines
		Attaching a pipeline server to your project
		Building a basic data science pipeline
	Summary
Chapter 5: Deploying ML Models as a Service
	Packaging and deploying models as a service
		Saving and uploading models to S3
		Updating the pipeline via model upload to S3
		Creating a model server for Seldon
		Deploying and accessing your model
	Autoscaling the deployed models
	Releasing new versions of the model
		Automating the model deployment process
		Rolling back model deployments
		Canary model deployment
	Securing model endpoints
	Summary
Chapter 6: Operating ML Workloads
	Monitoring ML models
	Installing and configuring Prometheus and Grafana
	Logging inference calls
	Optimizing cost
	Summary
	References
Chapter 7: Building a Face Detector Using the Red Hat ML Platform
	Architecting a human face detector system
	Training a model for face detection
		Deploying the model
		Validating the deployed model
	Installing Redis on Red Hat OpenShift
	Building and deploying the inferencing application
	Bringing it all together
	Optimizing cost for your ML platform
		Machine management in OpenShift
		Spot Instances
	Summary
Index
Other Books You May Enjoy