Test Drive New Features of Red Hat OpenShift AI

Introduction

Welcome!

In this lab, you will start with the basics of OpenShift AI, and end with some of the newest features.

If you’ve never used OpenShift AI, don’t worry. We will start with a straightforward fraud detection model built with Tensorflow to learn the basics. Then we will progress to some new features of OpenShift AI that allow it to be a great platform for generative AI and large language models (LLMs). This will include distributed training and LLM model serving.

You will use an example fraud detection model to complete the following tasks:

  • Explore a fraud detection model by using Jupyter Notebooks.

  • Deploy the model by using OpenShift AI single model serving.

  • Refine and train the model by using automated pipelines.

Using an LLM (Google’s flan-t5-small), we will complete the following tasks:

  • Download the model from HuggingFace

  • Upload it into our Object Storage

  • Serve it with OpenShift AI’s Single Model serving