This workshop provides demonstrations of the effective usage of popular machine learning libraries on the HPC clusters at Princeton. It shows how to install each library as well as how to write Slurm scripts to take advantage of multi-threading and/or GPUs. The material covers PyTorch, TensorFlow, JAX, Julia, Spark, Rapids, Scikit-learn, R and Snap ML.
Tue May 18, 9am-12:30pm & Wed May 19, 9am-12:30pm
Virtual workshop (Zoom link will be emailed to accepted participants)
Registration is required
Dates: June 2, 8-10
Become a mentor
Della will soon have 20 nodes with 2 A100 GPUs per node.
TigerGPU will hopefully be replaced in the Fall.
AMD GPU node coming to Adroit in the coming weeks.
If you encounter any difficulties while working with the machine learning libraries on the HPC clusters then please send an email to cses@princeton.edu or attend a help session.
This workshop was created by Jonathan Halverson.