You are here

Machine Learning Workshop: Tensorflow / Keras for Visual Recognition Models

Overview

Where

29 Little W 12th St
New York, NY 10014
United States

When

Monday, November 12, 2018 -
3:00pm to 5:00pm

Details

Deploy to CoreML for Edge Computing

Image recognition is one of the most explosive growth areas in machine learning. Get a hands-on introduction at our workshop.

Use Cases:

Here's just a few of the exciting uses of machine learning:

- photo search and clustering (find specific subjects, find all photos with particular people)
- self-driving cars (recognizing road hazards, vehicles, signs)
- identifying image control points and applying dynamic filters for fun (Snapchat) or commerce (see how an item would look on you)
- fashion and art commerce (identifying similar styles you'd be interested in)
- medical diagnosis (analyze various types of scans)
- security (face or iris based authentication, license plate identification)

What You'll Learn:

This talk and tutorial will teach you how to train a Visual Recognition Classifier in Watson Studio using Tensorflow and Keras open-source libraries.

It's all on a cloud based lab so no need for any special hardware or local installation to take part.

Secondly, the talk will cover next steps on how to convert the Keras model to a CoreML model for edge computing right on an iPhone/iPad. No cloud or internet required!

Why is edge computing exciting? The response time is fast and data stays private for users not moving to the cloud and there's no need for a constant internet connection (or using up data plans).

You will see a demo of the final product in an iOS application, where the application is able to identify various objects trained in the custom model.

What You Need:

No previous programming experience is required as all the code will be walked through by instructors.

Just bring a charged laptop (and your charger just in case).

Trainers:

Nick Bourdakos (@bourdakos1), Developer Advocate, IBM

Nick is a developer advocate at IBM Watson in NYC. His expertise is in machine learning, mainly deep learning applied to Computer Vision problems. He started out as an Android developer, but also does Swift, Python and JavaScript / React development.

Helen Lam (@helennnsays), Developer Advocate, IBM

Helen is a developer advocate at IBM in NYC. Her expertise and background include building using Watson APIs. Her stack includes Ruby-on-Rails, and JavaScript and Node JS development.

Event Schedule:

3:00pm - 5:00pm Visual Recognition Workshop in Library Room

5:00pm - 5:30pm Post workshop Q&A with trainers and attendees in the main lounge

5:30pm - 6:30pm Networking, food and drinks

The workshop is over and we'd love for you to stay for our follow-up event on machine learning for driving team productivity!

6:30pm - 6:35pm Kickoff

6:35pm - 7:10pm Eric Typaldos, Founder and CTO of Hive - Using Machine Learning to Build Better Teams

7:10pm - 7:40pm TBA

7:40pm - 7:50pm Community Asks and Offers

7:50pm - 8:45pm Post event networking

Sponsors:

This event is made possible with the support of IBM Watson and Betaworks Studios.