The Luxembourg Competence Centre in High-Performance Computing (HPC), in collaboration with NVIDIA, is hosting online the AI for Science and Engineering Bootcamp during 2 half-days. The first part will be dedicated to theory, and the second part will focus on hands-on challenges on GPU accelerators of the MeluXina supercomputer.
For whom?
Both current or prospective users of large hybrid CPU/GPU clusters which develop HPC and AI applications that could benefit from GPU acceleration are encouraged to participate!
What will you learn and how?
During this online Bootcamp, participants will learn how to apply AI tools, techniques, and algorithms to real-life problems. Participants will be introduced to the critical concepts of Deep Neural Networks, how to build Deep Learning models, and how to measure and improve the accuracy of their models. Participants will also learn essential data pre-processing techniques to ensure a robust machine-learning pipeline. The Bootcamp is a hands-on learning experience where mentors guide you.
Learning outcomes
After this course, participants will be able to:
- Apply Deep Convolutional Neural Networks for science and engineering applications
- Understanding the Classification (multi-class classification) methodology in AI
- Implement AI algorithms using Keras (e.g. TensorFlow)
- Use an efficient usage of the GPU for AI algorithms (e.g. CNN) with handling large data set
- Run AI applications in the Jupyter notebook environment (and understand singularity containers)
Prerequisites
Priority will be given to users with basic experience with Python. No GPU programming knowledge is required.
GPU Compute Resource
Teams attending the event will be given access to the MeluXina supercomputer during the hackathon. To learn more about MeluXina, please consult the Meluxina overview and the MeluXina – Getting Started Guide.
