The Conference for Machine Learning Innovation

An Introduction to Deep Learning with Keras – train your own Image Classifier with Deep Neural Networks [SOLD OUT]

This talk originates from the archive. To the CURRENT program
Join the ML Revolution! Register until September 26th: ✓Arduino Starter Kit or C64 Mini for free ✓Save up to € 300 ✓ 10% Team Discount Register Now
Monday, June 17 2019
09:00 - 17:00
Booking note:
Keras Workshop

In this workshop, you will learn how to create a neural network that can classify images into their respective classes (Image Classification). For this, we’ll be using the Fashion-MNIST dataset. While starting with a very basic, shallow network and the underlying ideas, we gradually add more complexity and introduce convolutional layers to improve the performance of our model. At the end, you will have a solid understanding of the underpinnings of deep learning and have all the tools needed to start applying these to your own projects.

We’ll start with a general introduction into the Deep Learning paradigm. This involves some very basic linear algebra (like matrix multiplication), but nothing too difficult. We’ll quickly take a look at what the backpropagation algorithm does and how it can be used to train a neural network. After this we’ll dive straight into the Deep Learning frameworks we’ll be using: Tensorflow and Keras.

With a series of assignments we then progressively start making a more and more capable neural network to solve an immage classification problem using a relatively small dataset of 60’000 labeled images. At the end we introduce Convolutions, regularization, dataloaders and custom callbacks to make the training process easier and more transparent.

1. Introduction to Deep Learning: foundations

2. Introduction to TensorFlow and Keras

3. Introduction to Google Cloud platform and training on GPU

4. Build your first neural network using Keras

5. Intro to Convolutional neural networks & making the classifier better

For this workshop we only require participants to have a basic knowledge of Python in order to work on the hands-on coding exercises. If you do not have any coding experience you are still welcome to join the workshop as all the solutions will be shared at the end. For the workshop we will use Google’s colab notebooks that come with GPU acceleration and already have all prerequisites installed.

Every participant should bring their own laptop and consider if VPN restrictions might block the connection to Google Colab.

Behind the Tracks