Convolutional Neural Networks for Image Processing (Cape Town)

Convolutional Neural Networks (CNNs) are a kind of neural network - a so called “deep learning” model - that is at the cutting edge of computer vision applications.

Alex Conway (founder & CTO of NumberBoost) will present an introduction to the topic as well as tutorial in the form of a python notebook on using CNNs for image classification.

We will cover:

• a brief introduction to neural networks

• the architecture of a convolutional neural network

• the different layer types and operations

• how to fine tune an existing model to a new task

• practical tips* getting the model running on an AWS GPU instance

• some state of the art results and architectural tweaks