Module 4 - Deep Learning with MATLAB

on Friday, 04/12/2024, 2:00 pm - 5:00 pm, CET

Speaker: Franziska Albers, Academia Team Member, MathWorks

This tutorial will focus on how you can use the GPUs available on the ZIH cluster for training of deep neuronal networks in MATLAB. We will discuss best practices for deep learning training, and you will learn where and how to store your data for best performance. We will also cover the workflow for hyperparameter optimization.

We will not cover the basic of deep learning or how to design a neural network in MATLAB. The presentation will be held online and in English. Registration is required.

Please note: We highly recommend joining the second MATLAB Tutorial Part:
Scaling Parallel Computing with MATLAB to HPC @ TU Dresden before you join this one. This tutorial will utilize knowledge and setup from that workshop to submit deep learning jobs to the cluster.

Prerequisites

• A ZIH cluster user account. This account will be created for you when you register.
• Local installation of MATLAB R2023b including
  • Parallel Computing Toolbox
  • Image Processing Toolbox
  • Deep Learning Toolbox
  • Statistics and Machine Learning Toolbox
• Basic knowledge of deep learning (if you need a refresher, please complete the Deep Learning Onramp).

Learning Objectives

Participants will be able to apply deep learning concepts to their research and engineering problems.

Registration

Link: https://event.zih.tu-dresden.de/nhr/matlab-4
Registration is closing on 04/10/2024. The NHR tutorial is limited to 60 participants.
You will receive the access data shortly before the event by email to your registered email address.