Intuitive deep-learning crash course
These are the slides from Workshop: Building your first Convolutional Neural Network
This introductory workshop will show you – a very beginner – how to build and train your very first Convolutional Neural Network. We will go over basic deep learning theory and build a network that can classify grayscale images of fashion items into various categories. The workshop aims to show you basic practical skills and tools needed to start applying deep learning to your own image classification problems right away.
The workshop will consist of two parts. The first part will provide a brief overview of deep learning theory, while the second part will show you how to build a network for an image classification problem. We will show how to install necessary dependencies on a Windows machine, look at the chosen image dataset, and set everything up for appropriate training and evaluation. We will build, train and evaluate our first basic Convolutional Neural Network (CNN). The workshop is intended for beginners and aims at providing a basic showcase of deep learning in a compact session.
Keywords: Python, Tensorflow, Keras, Fashion MNIST