[ONLINE] Introduction to Deep Learning with PyTorch

16 Sept 2025, 09:00 → 17 Sept 2025, 15:30 Europe/Prague

ZOOM (ONLINE)

Annotation

This training introduces participants to PyTorch in an HPC environment, covering deep learning, fine-tuning, and testing neural network models, as well as implementing concepts such as distributed data parallelism. Designed for researchers and developers, the course includes hands-on sessions to enhance practical skills.

The knowledge and skills acquired in this course are highly relevant for professionals working in companies across various sectors. By learning how to train and deploy neural network models in a high-performance computing environment, participants will be equipped to tackle data-intensive tasks such as automated quality inspection, customer behaviour prediction, demand forecasting, or sensor data classification. These techniques can support innovation and enhance efficiency in various fields, including manufacturing, healthcare, finance, agriculture, and logistics. The practical experience gained during hands-on sessions ensures that attendees leave with applicable tools for developing scalable, AI-driven solutions tailored to their industry’s needs.

Benefits for the attendees, what will they learn:

• Understand PyTorch’s Tensor library and neural networks at a high level in HPC
• Parallelizing the training of deep learning models across multiple GPUs and machines.
• Train and test neural network models
• Hands-on experience with lab exercises for practical skills

 

Participants will have access to the Karolina supercomputer for hands-on sessions, utilizing both CPU and GPU resources. Karolina, operational since 2021, is the most powerful supercomputer in the Czech Republic and ranks among Europe's top systems. It features a standard part with 720 nodes, delivering 11.6 PFlop/s for traditional HPC simulations, and an accelerated section comprising 72 servers, each equipped with 8 GPU accelerators, achieving up to 360 PFlop/s for AI computations. 

Level

70% beginner, 30% intermediate

Language

English

Prerequisites

Experience with programming in Python and a decent level of mathematics.

Technical requirements: 

• Python and its dependencies.
• Jupyter Notebook for interactive coding.

Tutor

Ghaith Chaabane, Ph.D., is a Researcher at the Advanced Data Analysis and Simulation Laboratory within the IT4Innovations National Supercomputing Centre.

Acknowledgements

LUMI AI Factory is funded jointly by the EuroHPC Joint Undertaking, through the European Union's Connecting Europe Facility and the Horizon 2020 research and innovation programme, as well as Finland, the Czech Republic, Poland, Estonia, Norway, and Denmark.

 

This course was supported by the Ministry of Education, Youth and Sports of the Czech Republic through the e-INFRA CZ (ID:90254).

 

All presentations and educational materials of this course are provided under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.

Tuesday, 16 September

09:00 → 09:30 Welcome and Introduction

Overview of the workshop objectives.

09:30 → 10:30 Introduction to PyTorch, Neural Network Regression with Pytorch

Learn the basics of PyTorch and build a simple regression model using neural networks.

10:30 → 10:45 Coffee Break

10:45 → 12:00 PyTorch Neural Network Classification

Implement a neural network for classification tasks with hands-on practice.

12:00 → 13:00 Lunch Break

13:00 → 14:00 PyTorch Computer Vision

Explore computer vision tasks using convolutional neural networks (CNNs) in PyTorch.

14:00 → 14:45 PyTorch RNNs

Understand and apply Recurrent Neural Networks for sequence modeling.

14:45 → 15:00 Coffee Break

15:00 → 16:00 PyTorch Transfer Learning

Learn to adapt pre-trained models to new tasks using transfer learning techniques.

16:00 → 16:30 PyTorch Benchmarking

Implement data parallelism for scalable training on multiple GPUs or nodes.

16:30 → 17:00 Q&A and Wrap-up

Wednesday, 17 September

09:00 → 09:30 Recap of Day 1

Summary of key concepts covered on Day 1

09:30 → 10:30 PyTorch Custom Datasets

Create and use custom datasets for training deep learning models in PyTorch.

10:30 → 10:45 Coffee Break

10:45 → 12:00 Pytorch Distributed Data Parallel

Implement data parallelism for scalable training on multiple GPUs or nodes.

12:00 → 13:00 Lunch Break

13:00 → 14:30 PyTorch Model Deployment

Deploy PyTorch models to the internet and make them publicly accessible.

14:30 → 15:00 Q&A and Wrap-up