DataSciPy: Nvidia GPU Cloud

Train your models on NVIDIA GPUs from the comfort of your laptop!

One would love to accelerate Neural Network experiments using GPUs, it was a nightmare to setup all the required software, but it’s not that difficult anymore. Docker containers provided by the NVIDIA’s GPU Cloud (NGC) technology come with everything needed (Libraries, CUDA etc.) and connected ready for the user, one can set them up either on cloud platforms or on personal computers.

In this talk, Thrilok will walk us through the steps for setting up the NGC containers, manipulating the images and their instances on a Linux desktop. Depending on the availability of time, the entire process to have the system ready on Google Cloud will be demonstrated for the attendees to follow along and set up their own system.

Thrilok is a recent graduate of Michigan State University from Master of Science in Industrial Mathematics program, trained in Statistics, Mathematics and Machine Learning. He is currently working as an Associate Data Scientist at Zuben AI developing prototype models for challenging business questions.