Our two days online course on high-performance machine learning provides the necessary skills to train neural networks and extract the most relevant information from datasets.
High-Performance Deep Learning
Regularity
Every 4 months
Trainers
Caspar van Leeuwen
Monica Rotulo
What will you learn?
During our hands-on sessions you will have the opportunity to work on our high-performance systems with different types of data, and learn how to tune your model to obtain optimal results in the most efficient way. The first day will focus on basic deep learning knowledge and skills. The second day will focus on training neural networks on high-performance computing clusters.
On day 1 of this course you will:
- Understand the fundamental theories of machine learning and the intuitions/ideas behind the algorithms
- Work with a high-level machine learning API (Keras)
- Explore hyperparameter space to improve a neural network
- Understand the pitfalls of classic machine learning algorithms
On day 2 of this course you will learn:
- how to set up your software environment, and why the preinstalled software modules are useful
- how the file I/O might limit your training speed, and how to overcome that
- about the technical capabilities of modern day CPUs and GPUs (reduced precision datatypes, vector/matrix instructions)
- how to find bottlenecks in your code through creating a (PyTorch) profile
- how to use multiple CPUs or GPUs in a single training (parallel computing for deep learning)
For Whom?
Everyone interested in getting familiar with machine learning at scale, from the beginning up to more advanced topics
Prerequisites
- Basic knowledge on statistics
- Basic knowledge on linear algebra
- Basic knowledge on Python programming. Some experience with the use of Jupyter Notebooks is desirable, but not essential.
Basic knowledge on parallel computing is helpful, but not required.
Topics
Day 1:
- Introduction to Deep Learning
- Using the PyTorch framework
- Fully connected networks, Convolutional networks, Autoencoders (time permitting)
Day 2:
- Software installations on HPC systems
- Packed file formats for Machine Learning
- Parallel computing for deep learning
- Hardware (e.g. Tensor cores) and software features (e.g. low level libraries for deep learning) to accelerated deep learning
- Profiling PyTorch with TensorBoard
Costs
Participation is free of charge
The language of instruction is English
Do you want to participate?
In our agenda you can see all the upcomping trainings and events