Production ML pipelines with TensorFlow Extended (TFX)

Aurélien Geron (Kiwisoft)

9:00am—5:00pm Monday, October 28—Tuesday, October 29

Location: Great America Meeting Room 3

Average rating:

(5.00, 2 ratings)

Download slides (PDF)

Participants should plan to attend both days of training course. Note: to attend training courses, you must be registered for a Platinum or Training pass; does not include access to tutorials on Tuesday.

Aurélien Géron dives into creating production ML pipelines with TensorFlow Extended (TFX) and using TFX to move from ML coding to ML engineering. You'll walk through the basics and put your first pipeline together, then learn how to customize TFX components and perform deep analysis of model performance.

What you'll learn, and how you can apply it

Learn to use TFX to create production ML pipelines

Who is this presentation for?

You're a data scientist or ML engineer, and you want to move trained models to production on servers, mobile applications, or JavaScript.
You're a DevOps or ML ops engineer, and you need to create and maintain a production ML platform.
You're a researcher working with large datasets, and you need consistency and repeatability for resource-intensive workloads.

Prerequisites:

A working knowledge of machine learning and software development in Python

Hardware and/or installation requirements:

A Linux or macOS laptop, or access to a remote system

TFX is an end-to-end platform for deploying production ML pipelines. When you’re ready to move your models from research to production, use TFX to create and manage a production pipeline. You can deploy models to servers, mobile applications, or JavaScript with TFX.

Outline

Day 1

Issues and approaches in production software deployments and machine learning
TFX basic overview: Libraries, components, and metadata
Hands-on workshop: Your first pipeline—TFX on-premises with Airflow
Hands-on exercise: Developing custom components
Hands-on workshop: Alternate pipelines—A/B testing pipeline architecture

Day 2

Data wrangling with TFX: Finding and fixing problems with TensorFlow Data Validation (TFDV)
Hands-on exercise: Model understanding problem—TensorFlow Model Analysis (TFMA) and What-if
Deployment targets: Lite, JavaScript, and Serving
Hybrid cloud and on-premises deployments

About your instructor

Aurélien Géron is a machine learning consultant at Kiwisoft and author of the best-selling O’Reilly book Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow. Previously, he led YouTube’s video classification team, was a founder and CTO of Wifirst, and was a consultant in a variety of domains: finance (JPMorgan and Société Générale), defense (Canada’s DOD), and healthcare (blood transfusion). He also published a few technical books (on C++, WiFi, and internet architectures), and he’s a lecturer at the Dauphine University in Paris. He lives in Singapore with his wife and three children.

Conference registration

Get the Platinum pass or the Training pass to add this course to your package.

Comments on this page are now closed.

Comments

Kevin Shaw | CTO

10/25/2019 4:13am PDT

Are there any other packages we should pre-install? Apache Airflow? Python 3.7?

Sophia DeMartini | Senior Speaker Manager

10/24/2019 9:29am PDT

HI Melanie – the instructor sent me this answer:
The only prerequisite is to have a laptop with Chrome installed, and optionally Docker as well (will be used as a fallback in the unlikely event that there’s an Internet connection issue).

So, yes, a Windows laptop will work, as long as it has Chrome installed (and the browser needs to be installed BEFORE the attendee arrives onsite).

Melanie Rezac | Customer service

10/24/2019 6:39am PDT

Can the training be done on a Windows laptop?