NVIDIA’s Deep Learning Workshop
A two days deep learning instructor-led workshop on data and model parallelism organized by TU/e Supercomputing Center and hosted by NVIDIA. This is a unique opportunity for TU/e students and employees to deepen their expertise in scaling and deploying deep learning models on advanced NVIDIA hardware.
This workshop is open to all TU/e students and employees. Seats are limited, so make sure to register early and secure your spot at https://forms.office.com/e/UkELx9p9HG
Practical Information
- 💻 Bring your own laptop
- Lunch included
Prerequisites Day 1, Data Parallelism
- Experience with deep learning training using Python
Prerequisites Day 2, Model Parallelism
- Good understanding of PyTorch
- Solid knowledge of deep learning and data parallel training concepts
- Practice with deep learning and data parallelism (optional but helpful)
Questions? Contact supercomputing@tue.nl
Schedule
- Wed21JanData Parallelism – How to Train Deep Learning Models on Multiple GPUsLearn how to accelerate deep learning training by leveraging multiple GPUs efficiently.09:00 - 17:00 Neuron building | Room number 0.356
- Thu22JanModel Parallelism – Building and Deploying Large Neural NetworksDiscover techniques for training and deploying large-scale neural networks using model parallelism.09:00 - 17:00 Neuron building | Room number 0.356