This short course (2.5 hours) introduces researchers to Graphcore Intelligence Processing Units (IPUs) on the ACES cluster, a composable accelerator testbed at Texas A&M High Performance Research Computing. The instructor will demonstrate the execution of models from different frameworks on the IPU system, and participants will gain practical experience in converting TensorFlow and PyTorch code into IPU code through hands-on exercises.
Prerequisites: Current ACCESS ID; basic Linux/Unix skills; basic understanding of machine learning concepts, neural networks, and deep learning; familiarity with deep learning frameworks TensorFlow and/or PyTorch.
Read more at https://hprc.tamu.edu/training/aces_ipus.html