Presented By O’Reilly and Cloudera
Make Data Work
March 5–6, 2018: Training
March 6–8, 2018: Tutorials & Conference
San Jose, CA

Deploying deep learning with TensorFlow

Ron Bodkin (Google), Brian Foo (Google)
1:30pm–5:00pm Tuesday, March 6, 2018
Data engineering and architecture
 Location: LL21 B
Average rating: ***..
(3.00, 2 ratings)
Download slides (PDF)

Who is this presentation for?

  • Data engineers, analytics operations, and data scientists involved with production

Prerequisite knowledge

  • A basic understanding of the Linux system and scripting

Materials or downloads needed in advance

  • A laptop with virtual machine program, such as VMware, installed and the ability to log in to a public cloud environment.
  • A GitHub account (You'll get access to the course repository prior to the tutorial.)
  • Make sure to bring a laptop with the Chrome browser installed, and that your laptop can access the following two websites:
    • qwiklabs.com
    • cloud.google.com

What you'll learn

  • Learn best practices and approaches to deploy and scale deep learning models in production (and their trade-offs)

Description

TensorFlow and Keras are popular libraries for machine learning because of their support for deep learning and GPU deployment. Join Ron Bodkin and Brian Foo to learn how to execute these libraries in production with vision and recommendation models and how to export, package, deploy, optimize, serve, monitor, and test models using Docker and TensorFlow Serving in Kubernetes.

Topics include:

  • Deep learning model production considerations: Processor type, batching, and scheduling
  • Introduction to environment and example models
  • Serving basics: An overview of TensorFlow Serving and exporting models for serving and testing
  • Robust deployment: The fundamentals of Docker and Kubernetes and dockerizing and deploying models with Kubernetes
  • Scheduling and sharing models: How to support multiple models and monitor them, whether running on a CPU, GPU, or TPU
  • Optimizing models with the XLA compiler
Photo of Ron Bodkin

Ron Bodkin

Google

Ron Bodkin is a technical director on the applied artificial intelligence team at Google, where he provides leadership for AI success for customers in Google’s Cloud CTO office. Ron engages deeply with Global F500 enterprises to unlock strategic value with AI, acts as executive sponsor with Google product and engineering to deliver value from AI solutions, and leads strategic initiatives working with customers and partners. Previously, Ron was the founding CEO of Think Big Analytics, a company that provides end-to-end support for enterprise big data, including data science, data engineering, advisory, and managed services and frameworks such as Kylo for enterprise data lakes. When Think Big was acquired by Teradata, Ron led global growth, the development of the Kylo open source data lake framework, and the company’s expansion to architecture consulting; he also created Teradata’s artificial intelligence incubator.

Photo of Brian Foo

Brian Foo

Google

Brian Foo is a senior software engineer for Google Cloud working on applied artificial intelligence, where he builds demos for Google Cloud’s strategic customers and creates open source tutorials to improve public understanding of AI. Previously, Brian worked at Uber, where he trained machine learning models and built a large-scale training and inference pipeline for mapping and sensing/perception applications using Hadoop and Spark, and headed the real-time bidding optimization team at Rocket Fuel, where he worked on algorithms that determined millions of ads shown every second across many platforms such as web, mobile, and programmatic TV. Brian holds a BS in EECS from UC Berkeley and a PhD in EE telecommunications from UCLA.