Getting started with Deep Learning on Supercomputers
Description
We have developed a comprehensive tutorial for Deep Learning practitioners and/or researchers who intend to make the transition from single-GPU network training on a workstation, to multi-GPU and multi-node training on supercomputers. This free-to-use and publicly accessible tutorial walks the user through all the necessary steps required from creating and setting up an account on the JSC supercomputers, to successfully running code samples on these machines.
The tutorial is available here.
In addition, we teach “Deep Learning on Supercomputers” as part of the biannually offered training course “Introduction to the programming and usage of the supercomputer resources at Jülich”.
Our collaboration partners
Following is a list of our collaborators:
Related projects at the Simulation Lab Neuroscience
- hpc4neuro Python library
- Deep Learning software for Supercomputers
- Deep Learning for segmentation of 3D-PLI images
Simlab Contact
Last Modified: 19.05.2022