Many people think of training models when they hear "data science" or "machine learning." Training and models are certainly important, but making intelligent enterprise software involves many other tasks and technologies. These include gathering and processing data from disparate, voluminous sources; testing and comparing different algorithms; deploying models into production applications; continuous monitoring and updating of models, etc.
Enter our AI/ML platform cloud service: Red Hat OpenShift Data Science. OpenShift Data Science provides a platform where developers can easily collaborate with data scientists to develop, deploy, and monitor models. Find out how it can improve your machine learning projects, and explore hands-on resources to help you get up to speed quickly with natural language processing, Jupyter notebooks, and more.
What is Red Hat OpenShift Data Science?
OpenShift Data Science is a comprehensive environment based in the Red Hat OpenShift cloud service. It integrates the core IDE where data scientists train models—Jupyter notebooks—with important model development frameworks such as TensorFlow and PyTorch, as well as key open source partner technologies.
Using this service, data scientists, data engineers, and application developers can collaborate across the full life cycle of intelligent applications. The pre-integration in OpenShift Data Science relieves the burden of wiring these technologies together on your own. This lets you focus on creating differentiated value through machine learning with security, reliability, and performance at enterprise scale.
Technically, OpenShift Data Science is a sort of meta-operator that sits above other Kubernetes Operators and combines them into a coherent, integrated environment. OpenShift Data Science partner technologies today include:
- Anaconda Commercial Edition for secure distribution and package management;
- IBM Watson Studio for building and managing models at scale and for AutoML;
- Intel OpenVINO and oneAPI AI analytics toolkits for optimizing and tuning models;
- Seldon Deploy for deploying, managing, and monitoring models; and
- Starburst Galaxy for data integration.
Additionally, support for NVIDIA accelerated computing is coming very soon.
Try OpenShift Data Science
Interested in trying Red Hat OpenShift Data Science? What are you waiting for? Here are two ways to get started.
Build a natural language processing (NLP) application in the sandbox
We currently have a field trial of Developer Sandbox for Red Hat OpenShift where you can try OpenShift Data Science for free! Our new use case activity walks you through all the steps needed to develop, train and deploy a Natural Language Processing model using Red Hat OpenShift Data Science.
Use Jupyter notebooks and more with learning paths
To accompany our use case activity, we have also created two learning paths that show you how to:
- Launch Red Hat OpenShift Data Science.
- Get started with OpenShift Data Science learning materials, including how-to resources, tutorials, and quick starts.
Each learning path is composed of multiple learning resources. For example, the Launch Red Hat OpenShift Data Science learning path includes a dedicated resource to guide you through each of the following activities:
- Navigating the OpenShift Data Science dashboard.
- Launching JupyterHub.
- Exploring the JupyterLab environment.
- Using a Jupyter notebook.
This way, you can pick the learning path or resource that you need. Already know how to use a Jupyter notebook? You can skip that learning resource and move to the next one. It's that easy.
Continue your data science journey
Stay tuned, as each month we will release new learning paths. From creating PyTorch models to using NVIDIA GPUs, our learning paths will get you up and running in no time. These practical resources will show you how developing and deploying intelligent enterprise software can be easy:
- Try OpenShift Data Science in the Developer Sandbox for Red Hat OpenShift.
- Get started with step-by-step learning paths.
- Learn more about Red Hat OpenShift Data Science.
Acknowledgments
The author would like to thank Michael Pieche, Will McGrath, Jeff DeMoss, and Karl Eklund for their contributions to this article.
Last updated: October 31, 2023