Register here if you plan to attend this session:
http://ibm.biz/RSEConf19With Deep Learning and AI becoming much more common in the world of research, a new set of skills need to be developed. These skill will include more extensive use of Python and management of various libraries and plug ins, as well as knowledge of common Machine Learning and Deep Learning frameworks. Software like Tensorflow, PyTorch, and Caffe are becoming more prevalent, with interaction via Python and Jupyter notebooks.
This session is an introduction to the hardware and software products available from IBM Systems to support the development and deployment of your Machine Learning and Deep Learning workloads. We will focus on the capabilities and tools available, as well as the common use cases, and how best to deploy and utilise these tools for Research. This will include: The AC922 server, based on the POWER9 processor with NVLink for high bandwidth, Low latency connection to NVIDIA Tesla V100 GPUs. No charge software bundle incorporating optimised versions of the common Deep Learning frameworks (Tensorflow, PyTorch, Caffe). Enterprise AI development tools including Watson Machine Learning Accelerator, Watson Studio, and H2O Driverless AI. PowerAI Vision training and inference offerings for rapid development of Deep Learning imaging jobs. Deployment methodologies - whole system, containerised, multi-tenancy, scheduler based. Unique capabilities of the IBM platform, such as Large Model Support and SnapML for accelerated Machine Learning.
