This course gives a practical introduction to deep learning, including common neural network architectures such as convolutional, recurrent and transformer models. Also covered are GPU computing and tools to train and apply deep neural networks for natural language processing, images, and other applications.

The course consists of lectures and hands-on exercises using PyTorch. Participants will get to run exercises on the GPU-accelerated LUMI – one of Europe’s fastest supercomputers!

The course will be hosted by Aalto Scientific Computing and held at the Aalto University campus in Espoo, Finland. A limited remote participation option via Zoom is also possible.

This course is organized by the LUMI AI Factory (LAIF). Therefore, the course is open and free of charge for participants from academia, industry, and public administration from all EU countries and Norway. Please register with your organization’s email address to prove your affiliation.

Learning outcome

After the course the participants should have the skills and knowledge needed to begin applying deep learning for different tasks and utilizing the GPU resources available at CSC for training and deploying their own neural networks.

Prerequisites

The participants are assumed to have working knowledge of Python and suitable background in data analysis, machine learning, or a related field. Previous experience in deep learning is not required, but the fundamentals of machine learning are not covered on this course. Basic knowledge of a Linux/Unix environment will be assumed.

Course contents

- Introduction to deep learning
- Using Jupyter Notebooks course environment
- Basics of deep learning and how to train a neural network
- Image data and convolutional neural networks
- Text data and recurrent and transformers neural networks
- Running jobs on the LUMI supercomputer
- Using multiple GPUs and multiple nodes
- Exercises using PyTorch