!!! This is a hands-on code lab, please bring your laptop !!!

We will use Google Codelab to learn how to build and train a neural network that recognises handwritten digits. Along the way, as you enhance your neural network to achieve 99% accuracy, you will also discover the tools of the trade that deep learning professionals use to train their models efficiently.

This code lab uses the MNIST (http://yann.lecun.com/exdb/mnist/) dataset, a collection of 60,000 labeled digits that has kept generations of PhDs busy for almost two decades. You will solve the problem with less than 100 lines of Python / TensorFlow code.

You will learn:

1) What is a neural network and how to train it
2) How to build a basic 1-layer neural network using TensorFlow
3) How to add more layers
4) Training tips and tricks: overfitting, dropout, learning rate decay...
5) How to troubleshoot deep neural networks
6) How to build convolutional networks

What you'll need:

1) Python 2 or 3 (Python 3 recommended)
2) TensorFlow
3) Matplotlib (Python visualisation library).

The code is available at https://github.com/martin-gorner/tensorflow-mnist-tutorial

git clone or download from the above link.

Agenda

18:30 Doors open, networking over food & drinks

19:00 Demonstration - MySkillsApp Firebase & Angular app

19:30 Machine learning code lab

20:45 Questions & discussion

Thanks to Priocept (http://priocept.com) for sponsoring the event.