Introduction

• Kubeflow on OpenShift vs public cloud managed services

Overview of Kubeflow on OpenShift

• Code Read Containers
• Storage options

Overview of Environment Setup

• Setting up a Kubernetes cluster

Setting up Kubeflow on OpenShift

• Installing Kubeflow

Coding the Model

• Choosing an ML algorithm
• Implementing a TensorFlow CNN model

Reading the Data

• Accessing a dataset

Kubeflow Pipelines on OpenShift

• Setting up an end-to-end Kubeflow pipeline
• Customizing Kubeflow Pipelines

Running an ML Training Job

• Training a model

Deploying the Model

• Running a trained model on OpenShift

Integrating the Model into a Web Application

• Creating a sample application
• Sending prediction requests

Administering Kubeflow

• Monitoring with Tensorboard
• Managing logs

Securing a Kubeflow Cluster

• Setting up authentication and authorization

Troubleshooting

Summary and Conclusion