In this three-hour, hands-on workshop, the NeSI team will introduce you to the basics of convolutional neural networks (CNNs) for image processing, using the Tensorflow Python package. The team will work through building a simple neural network to classify images and reuse larger, pre-trained models to tackle more challenging tasks. Finally, you will convert your code into a non-interactive SLURM job to leverage GPU resources and get results faster and more efficiently.

This is a hands-on, online event

The workshop will address the following topics:

• Convolutional Neural Networks (CNNs)
• Tensorflow
• Image Analysis
• Classification
• Transfer Learning
• SLURM
• Other necessary details required to work on NeSI

Who is this event for

To attend this workshop, you will need to be familiar with the Python language or be competent in another analytical language (C, Matlab, Julia, R) and know how to enter commands in a terminal to navigate between folders and run programs. You should be familiar with concepts of predictive models as well. If you want to brush up on your Python skills beforehand please consider reviewing this programming with python material.

Setup requirements

You will need to have a NeSI account in order to fully participate in the workshop. Those that register will be sent instructions on how to do this before the event.

If you have any questions or would like more information about this session, please email training@nesi.org.nz.