Making Open Work
May 8–9, 2017: Training & Tutorials
May 10–11, 2017: Conference
Austin, TX

Learning TensorFlow (Day 2)

Tom Hope (Independent), Itay Lieder (Independent), Yehezkel Resheff (Independent)
Location: Meeting Room 7

Deep learning has emerged as the premier technology in all things AI. With open source frameworks making this technology widely available, deep learning is rapidly becoming a must-know for anybody in big data and machine learning.

Tom Hope, Itay Lieder, and Yehezkel Resheff, the authors of O’Reilly’s upcoming book Learning TensorFlow: A Guide to Building Deep Learning Systems, walk you through the fundamental concepts of deep learning and explain how to build production-ready AI systems with TensorFlow, Google’s open source deep learning framework, which is currently the leading software framework for building machine intelligence systems with deep learning.


Day 1

9:00am – 10:00am: Intro

  • Course logistics and introduction
  • What is machine learning?
  • From ML to DL
  • What is TensorFlow, and what is it good for?

10:00am – 10:30am: Getting started

  • Installing TensorFlow
  • Running the “Hello, World” example and a couple other simple examples

10:30am – 11:00am: Morning break

11:00am – 12:30pm: Internals—Computation graphs

  • What are computation graphs?
  • TensorFlow’s computation graph model
  • Hands-on exercise

12:30pm – 1:30pm: Lunch

1:30pm – 3:00pm: CNNs

  • What are convolutional neural networks?
  • Building your first CNN in TensorFlow
  • Hands-on exercise

3:00pm – 3:30pm: Afternoon break

3:30pm – 5:00pm: Visualization with TensorBoard

  • Lecture and hands-on exercise

Day 2

09:00am – 10:30am: Abstraction layers-slim

  • Lecture and hands-on exercise

10:30am – 11:00am: Morning break

11:00am – 12:30pm: Pretrained vision models

  • Lecture and hands-on exercise

12:30pm – 1:30pm: Lunch

1:30pm – 3:00pm: TensorFlow for text—Part I

  • Intro
  • Basic models (RNN and word embeddings)

3:00pm – 3:30pm: Afternoon break

3:30pm – 4:00pm: TensorFlow for text—Part II

  • RNNs continued
  • Hands-on exercise

4:00pm – 4:40pm: Additional topics

  • Queues, threads, and reading data
  • Distributed TensorFlow
  • Serving models

4:40pm – 5:00pm: Final remarks

  • Recap
  • Where to go from here to further learn the subject?
  • Q&A
Photo of Tom Hope

Tom Hope


Tom Hope is an applied machine-learning researcher and data scientist. Tom has an extensive background as a senior data scientist for large, international corporations, where he has led data science and deep learning R&D across multiple domains, including web mining, text analytics, computer vision, sales and marketing, the IoT, financial forecasting, and large-scale manufacturing. Previously, Tom was at ecommerce startup Tapingo in its early days, where he led data science R&D. He has also served as a data science consultant for major international companies and startups. Tom’s academic research and publications in computer science, data mining, and statistics revolve around machine learning, deep learning, NLP, weak supervision, and time series.

Photo of Itay Lieder

Itay Lieder


Itay Lieder is a computational neuroscience researcher who uses advanced machine-learning tools to understand and model auditory perception. Itay has industry experience as a data scientist in deep learning R&D, focusing on text mining and NLP.

Photo of Yehezkel Resheff

Yehezkel Resheff


Yehezkel Resheff is an applied machine-learning researcher with years of experience in both academic and industrial settings. Currently, Yehezkel is working on bringing deep learning into new domains. Previously, Yehezkel was an applied researcher at Intel and Microsoft. His graduate work focused on machine learning for wearable devices.