NVIDIA DLI WORKSHOP FUNDAMENTALS OF DEEP LEARNING FOR COMPUTER VISION | SGInnovate
March 26
2018

Location

Pisces 1, Level 1, Resorts World Convention Centre Sentosa
Singapore Singapore 098269

NVIDIA DLI WORKSHOP FUNDAMENTALS OF DEEP LEARNING FOR COMPUTER VISION

Presented by NVIDIA, SGInnovate and National Supercomputing Centre

Organised by NVIDIA, SGInnovate and the National Supercomputing Centre, the NVIDIA Deep Learning Institute (DLI) offers hands-on training for developers, data scientists, and researchers looking to solve the world’s most challenging problems with deep learning. DLI is excited to announce this one-day practical Deep Learning workshop at SupercomputingAsia 2018 (SCA18).

Instructor: Oh Chin Lock

An experienced educator, Chin Lock has been with Temasek Polytechnic for more than 10 years. He has played an instrumental role in launching courses in the areas of Business Analytics and Big Data Management, and played a key role in building industry collaboration and staff capability in other key areas such as machine learning, agile software development, IoT and cyber security. Prior to joining Temasek Polytechnic, Chin Lock was in industry designing and building software products for both large enterprises and an e-commerce start-up. He has a keen interest in innovation and technology. Chin Lock holds a MSc in IP Management and a BSc (Hons) in Computer & Information Sciences. He is a senior member of the Singapore Computer Society where he also serves as ICT career mentor.

 

Important Notes to Participants:

  • Please register for this workshop on the SuperComputing Asia 2018 (SCA2018) website here, https://sca-2018.com/register/other/tutorial.
  • You must bring your own laptop, charger and adaptor (if required) to this workshop.
  • Create an account by going to https://nvlabs.qwiklab.com/ prior to getting to the workshop.
  • Make sure your laptop is set up prior to the workshop by following these steps:
    • Ensure WebSockets runs smoothly on your laptop by going to http://websocketstest.com/
    • 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 or trying a different browser. The labs will not run without WebSockets support.
  • Best browsers for the labs are Chrome, FireFox and Safari. The labs will run in IE but it is not an optimal experience.
  • Please remember to sign in to nvlabs.qwiklab.com using the same email address as for event registration, since class access is given based on the event registration list.

Abstract of the Tutorial:

The NVIDIA Deep Learning Institute (DLI) offers hands-on training for developers, data scientists, and researchers looking to solve the world’s most challenging problems with deep learning. DLI is excited to announce this one-day practical Deep Learning workshop at SupercomputingAsia 2018 (SCA18).

In this full-day workshop, you’ll learn the basics of deep learning by training and deploying neural networks. You’ll learn how to:

  • Implement common deep learning workflows, such as image classification and object detection.
  • Experiment with data, training parameters, network structure, and other strategies to increase performance and capability.
  • Deploy your neural networks to start solving real-world problems. Upon completion, you’ll be able to start solving problems on your own with deep learning.

Agenda

08:30 Registration

09:00 Deep Learning Demystified (lecture)

09:45 Image Classification with DIGITS (hands-on lab)

10:30 Tea Break

11:00 Image Classification with DIGITS (hands-on lab) [Continued]

12:30 Lunch Break

13:30 Approaches to Object Detection with DIGITS (hands-on lab)

15:15 Neural Network Deployment with TensorRT (hands-on lab)

16:00 Tea Break

16:30 Neural Network Deployment with TensorRT (hands-on lab) [Continued]

17:00 Closing Comments and Questions

*Agenda is subjected to change Content

Level: Beginner

Training Syllabus:

Lab 1: Image Classification with DIGITS

  • Deep learning enables entirely new solutions by replacing hand-coded instructions with models learned from examples. Train a deep neural network to recognize handwritten digits by:
  • Loading image data to a training environment.
  • Choosing and training a network.
  • Testing with new data and iterating to improve performance.

On completion of this Lab, you will be able to assess what data you should be training from.

Lab 2: Object Detection with DIGITS

Many problems have established deep learning solutions, but sometimes the problem that you want to solve does not. Learn to create custom solutions through the challenge of detecting whale faces from aerial images by:

  • Combining traditional computer vision with deep learning.
  • Performing minor “brain surgery” on an existing neural network using the deep learning framework Caffe.
  • Harnessing the knowledge of the deep learning community by identifying and using a purpose-built network and end-to-end labeled data.

Upon completion of this lab, you will be able to solve custom problems with deep learning.

Lab 3: Neural Network Deployment with DIGITS and TensorRT

Deep learning allows us to map inputs to outputs that are extremely computationally intense. Learn to deploy deep learning to applications that recognize images and detect pedestrians in real time by:

  • Accessing and understanding the files that make up a trained model.
  • Building from each function’s unique input and output.
  • Optimising the most computationally intense parts of your application for different performance metrics like throughput and latency.

Upon completion of this Lab, you will be able to implement deep learning to solve problems in the real world.

Pre-requisites:

  • No background in deep learning is required for this training.
  • Basic python understanding can be useful for some exercises.
  • The mathematical and theoretical aspects of deep learning will NOT be covered by this training and they’re not a requirement to complete the labs. Reading the Wikipedia page of Deep Learning would be a good start if you’re interested.

Topics: Artificial Intelligence / Deep Learning / Machine Learning / Robotics

You may also like the following: