OpenShift Commons Briefing #110: Containerizing TensorFlow Applications on OpenShift with Subin Modeel (Red Hat)
December 4, 2017 | by
OpenShift Commons Briefing Summary
Red Hat’s Subin Modeel talks about how you can get started using OpenShift for your TensorFlow application development. The briefing is a definitive guide to getting Tensorflow applications and workflows deployed and trained on OpenShift.
The briefing covers the following:
How to build TensorFlow binaries with S2I (Source-to-Image) feature on OpenShift.
How to create custom docker images for TensorFlow for running on GPU with OpenShift.
Using Jupyter notebook/jupyterab to write Tensorflow code, to serialize models to TensorFlow’s SavedModel format and to use OpenShift Source-to-image for creating models.
How to deploy the model as an API endpoint which can be consumed by applications.
Subin also demonstrates some fun applications on OpenShift that leverage the technology including:
MNIST handwritten digit recognition app on OpenShift for CPU & GPU
Inception app – CPU only
Neural style transfer app on OpenShift for CPU/GPU
Join the OpenShift Commons Machine Learning Special Interest Group:
The principal purpose of the Machine Learning on OpenShift Special Interest Group is to discuss, develop, and disseminate best practices for deploying and managing Machine Learning workloads and applications on OpenShift built using (but not limited to) TensorFlow, Apache Spark, and other Open Source ML/AI frameworks.
Learn More at the Next OpenShift Commons Gathering in Austin Dec 5th,
Red Hatters, CNCF/Kubernetes project leads, and numerous other members of the OpenShift Commons will be gathering together in Austin for the upcoming OpenShift Commons Gathering co-located with Kubecon at the Austin Convention Center.
Don't forget to leave your feedback and suggestions for each video on YouTube or in the comments section below. This will be incredibly important to shape this Special Interest Group and create sessions that fit the demands of all the OpenShift developers in the community.
About OpenShift Commons
OpenShift Commons is the place for organizations that are part of the OpenShift community to connect with peers and other related open source technology communities to communicate and collaborate across all OpenShift projects and stakeholders.
The Commons' goal is to foster collaboration and communication between OpenShift stakeholders to drive success for all members, and expand & facilitate points of connection between members for sharing knowledge and experience to help drive success for the platform and for participants: customers, users, partners, and contributors.
Introduction The Node Tuning Operator (NTO) is a Second Level Operator (SLO) that ships with OpenShift Container Platform (OCP) 4 by default. It has come a long way since its first release, and this ...