2020-07-14 ONLINE: Deep Learning and GPU programming using OpenACC @ HLRS Stuttgart

Overview

Learn how to train and deploy a neural network to solve real-world problems, how to generate effective descriptions of content within images and video clips and how to accelerate your applications with OpenACC.

The workshop combines lectures about Fundamentals of Deep Learning for Computer Vision and Multiple Data Types with a lecture about Accelerated Computing with OpenACC.

The lectures are interleaved with many hands-on sessions using Jupyter Notebooks. The exercises will be done on a fully configured GPU-accelerated workstation in the cloud.

Program (preliminary)

 1^st day: Fundamentals of Deep Learning for Computer Vision (9:00 - 18:00,  Dr. Juan Durillo Barrionuevo)

Explore the fundamentals of deep learning by training neural networks and using results to improve performance and capabilities.

During this day, 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.

2^nd day: Fundamentals of Deep Learning for Multiple Data Types  (9:00 - 18:00,  Dr. Juan Durillo Barrionuevo)

This day explores how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips.

Learn how to train a network using TensorFlow and the Microsoft Common Objects in Context (COCO) dataset to generate captions from images and video by:

• Implementing deep learning workflows like image segmentation and text generation
• Comparing and contrasting data types, workflows, and frameworks
• Combining computer vision and natural language processing

Upon completion, you’ll be able to solve deep learning problems that require multiple types of data inputs.

3^rd day: Guest Lecture & Fundamentals of Accelerated Computing with OpenACC (9:00 - 18:00)

On the 3rd day we start with a guest lecture (9:00 - 10:30) about Deep Neural Networks for Data-Driven Turbulence Models, using DL in CFD.
Dr.-Ing. Andrea Beck, Institute of Aerodynamics and Gas Dynamics, University of Stuttgart. The abstract can be found here. Slides of the talk.

In the second part (10:45 - 17:00, by Dr. Volker Weinberg) you learn the basics of OpenACC, a high-level programming language for programming on GPUs. Discover how to accelerate the performance of your applications beyond the limits of CPU-only programming with simple pragmas. You’ll learn:

• How to profile and optimize your CPU-only applications to identify hot spots for acceleration
• How to use OpenACC directives to GPU accelerate your codebase
• How to optimize data movement between the CPU and GPU accelerator

Upon completion, you'll be ready to use OpenACC to GPU accelerate CPU-only applications.

At the end (17:00 - 18:00) we provide an optional tutorial / quick practical recap on Python 3.X (Dr. Lorenzo Zanon)

4^th day: ML Examples and Methods on HLRS systems (9:00 - 16:00)

On the 4th day we cover Pre-processing of data, mathematical methods and machine learning as well as Deep learning on HLRS Systems.

• Focus on Pre-processing, Feature Engineering and Machine Learning:
  Stuttgart S-Bahn Example (Dr. Lorenzo Zanon, Li Zhong and Dennis Hoppe, HLRS)

• Data Compression of numerical data sets with the BigWhoop library (Patrick Vogler, HLRS in cooperation with EXCELLERAT)

• Deep learning example with mathematical background and using Tensorflow (Dr. Khatuna Kakhiani)

The first three days are organized in cooperation with LRZ (Germany) and Nvidia. All instructors are NVIDIA certified University Ambassadors. On the last day, you will learn more about data preparation and DL on the systems at HLRS.

Important information

After you are accepted, please create an account under courses.nvidia.com/join using the same email address as for event registration, since lab access is given based on the event registration list. Please be aware that for adminstrative reasons, after you register, Nvidia will use your email address to contact you for the final feedback of the workshop.

Course Fee

• Students without Diploma/Master: 40 EUR
• Students with Diploma/Master (PhD students) at German universities: 90 EUR
• Members of German universities and public research institutes: 90 EUR
• Members of universities and public research institutes within EU or PRACE member countries: 180 EUR.
• Members of other universities and public research institutes: 360 EUR
• Others: 960 EUR

NVIDIA Deep Learning Institute

The NVIDIA Deep Learning Institute delivers hands-on training for developers, data scientists, and engineers. The program is designed to help you get started with training, optimizing, and deploying neural networks to solve real-world problems across diverse industries such as self-driving cars, healthcare, online services, and robotics.

Rolf Rabenseifner phone 0711 685 65530, rabenseifner@hlrs.de
Lucienne Dettki phone 0711 685 63894, dettki@hlrs.de