Logo dotAI Open Nav

TensorFlow Training

TensorFlow and Deep Learning - without a PhD

Our Gold partner Google Cloud Platform is providing a TensorFlow training on April 26th.



Martin Görner

Martin Görner

Developer Relations at Google

Yufeng Guo

Yufeng Guo

Developer Advocate for Machine Learning at Google Cloud Platform


First session - Tensorflow and deep learning, without a PhD:

Codelab instructions: https://codelabs.developers.google.com/codelabs/cloud-tensorflow-mnist/#0

TensorFlow trainings

Duration: 3h

Programming language: Python

Abstract: Google has open-sourced its framework for machine learning and neural networks called Tensorflow. With this new tool, deep machine learning transitions from an area of research into mainstream software engineering. In this codelab, you will see how to choose the right neural network for your problem and how to make it behave. Familiarity with differential equations is no longer required. Instead, a couple of lines of Tensorflow Python, and a bag of "tricks of the trade" will do the job.
At the end of the lab, you will have coded a neural network that recognises handwritten digits with a 99.3% accuracy, with just 100 lines of Python/Tensorflow. Along the way, you will discover many standard deep learning techniques such as mini-batching, learning rate decay, dropout, convolutional networks and more. You will also learn how to implement them in Tensorflow.

Second session - Going Wide and Deep with TensorFlow in the Cloud

Duration: 3h

Programming language: Python

Abstract: Abstract: Training and deploying a custom machine learning model can be a long process. You want to spend more time working with your model and data rather than your infrastructure. Thankfully, the tooling that is available today can help make this a reality.
In this session, we will go on an adventure to build a machine learning model that combines the benefits of linear models with deep neural networks. You will gain some intuition about what is happening under the hood, and at each phase of the process, we will discuss the design decisions and tradeoffs available.
You will come away with a highly tuned machine learning model, and the skills to adapt it to solve machine learning problems on your own dataset.

Original Wide & Deep paper: https://arxiv.org/abs/1606.07792

Laptop requirements:

Python 3 is recommended for this lab. You will also need tensorflow and matplotlib installed.

To make sure everyone has the requirements installed, the only way to register to the workshop is to run this script!

Update 04-19-17: The workshop is now sold out but he still have tickets to dotAI here.