Deep Learning Hands-on Training Workshop

Join us for an afternoon of deep learning training hosted by the NVIDIA Deep Learning Institute (DLI).

Why deep learning is important to HPC

The mission of High-Performance Computing (HPC) is to tackle the world’s greatest challenges by building and computing models that represent the physical world. The traditional supercomputing approach uses expert-designed models, handcrafted by scientists using mathematical algorithms. The deep learning approach uses data-driven models, where software writes the algorithms; fine-tuning them with lots of data. Important scientific advancements have already been achieved using this approach, particularly in healthcare. Both approaches are critical to advancing HPC.  

 

DLI Workshop Schedule

12:30 pm		Attendees arrive
1:00 pm		Getting Started with Deep Learning
3:00 pm		Refreshment Break
3:30 pm		Deep Learning for Image Segmentation
5:00 pm		Workshop Concludes

DLI Lab Descriptions

Getting Started with Deep Learning

Presented by Jon Barker, NVIDIA Deep Learning Institute Approved Instructor and Deep Learning Research Scientist, NVIDIA

Deep learning is giving  machines near human levels of visual recognition capabilities and disrupting many applications by replacing hand-coded software with predictive models learned directly from data.  This lab introduces the machine learning workflow and provides hands-on experience with using deep neural networks (DNN) to solve a real-world image classification problem. You will walk through the process of data preparation, model definition, model training and troubleshooting, validation testing, and strategies for improving model performance. You’ll also see the benefits of GPU acceleration in the model training process. On completion of this lab, you will have the knowledge to use NVIDIA DIGITS to train a DNN on your own image classification dataset.

Prerequisites: Basic knowledge of data science and machine learning

Audience Level: Beginner

Deep Learning for Image Segmentation

Presented by Jon Barker, NVIDIA Deep Learning Institute Approved Instructor and Deep Learning Research Scientist, NVIDIA

There are a variety of important applications that need to go beyond detecting individual objects within an image and instead segment the image into spatial regions of interest. For example, in medical imagery analysis, it is often important to separate the pixels corresponding to different types of tissue, blood or abnormal cells so that we can isolate a particular organ. In this lab, we will use the DIGITS to train and evaluate an image segmentation network using a medical imagery dataset.  

Prerequisites: Basic knowledge of DIGITS

Audience Level: Intermediate

DLI Workshop Attendee Instructions

Create a qwikLABS account by going to https://nvlabs.qwiklab.com/ prior to getting to the conference.

• Make sure that WebSockets work for you by seeing under Environment, WebSockets is supported and Data Receive, Send and Echo Test all check Yes under WebSockets (Port 80).

• If there are issues with WebSockets, try updating your browser. Best browsers for qwikLABS are Chrome, FireFox and Safari. The labs will run in IE but it is not an optimal experience.