Deep Learning and GPU programming using OpenACC @ HLRS 2020
#
GROUP PICTURE
Foto: Volker Weinberg (LRZ)
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 course combines lectures about Fundamentals of Deep Learning for Computer Vision and Multiple Data Types with lectures about Accelerated Computing with OpenACC and DL on HLRS systems.
The lectures are interleaved with many hands-on sessions using Jupyter Notebooks. The exercises on the first 3 days will be done on a fully configured GPU-accelerated workstation in the cloud, on the last day on an HLRS system.
The first three days are organized by LRZ 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.
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.
TRAINING SETUP
To get started, follow these steps:
- Create an NVIDIA Developer account at http://courses.nvidia.com/join Select "Log in with my NVIDIA Account" and then '"Create Account".
- Make sure that WebSockets works for you:
- Test your Laptop at http://websocketstest.com
- Under ENVIRONMENT, confirm that '"WebSockets" is checked yes.
- Under WEBSOCKETS (PORT 80]. confirm that "Data Receive", "Send", and "Echo Test" are checked yes.
- lf there are issues with WebSockets, try updating your browser.
We recommend Chrome, Firefox, or Safari for an optimal performance. - Visit http://courses.nvidia.com/dli-event and enter the event code provided by the instructor.
- You're ready to get started. Please complete the survey at the end of the course to share your feedback.
1st day: Fundamentals of Deep Learning for Computer Vision
Lecturer: PD Dr. Juan Durillo Barrionuevo (LRZ)
AGENDA
09:00-09:20 Welcome and Intro
09:20-11:00 Introduction to Deep Learning and Training (I)
11:00-11:15 Coffee Break
11:15-12:00 Introduction to Deep Learning and Training (II)
12:00-13:00 Lunch Break
13:00-15:00 Deploying Neural Networks and Performance
15:00-15:20 Coffee Break
15:20 Group Picture
15:20-16:45 Beyond classification: Object Recognition
16:45-17:00 Q&A
SLIDES
FundamentalsDeepLearningComputerVision.pdf
2nd day: Fundamentals of Deep Learning for Multiple Data Types
Lecturer: PD Dr. Juan Durillo Barrionuevo (LRZ)
AGENDA
09:00-09:20 Welcome and Intro
09:20-12:00 Introduction to CNNs and Object Segmentation (I)
11:00-11:20 Coffee Break
11:20-12:00 Introduction to CNNs and Object Segmentation (II)
12:00-13:00 Lunch Break
13:00-15:00 Word Generation with RNNs
15:00-15:30 Coffee Break
15:30-16:45 Image Captioning by Combining RNNs and CNNs
16:45-17:00 Q&A
SLIDES
DOCUMENTATION
3rd day: Guest Lecture & Fundamentals of Accelerated Computing with OpenACC
Speaker: Dr.-Ing. Andrea Beck, Institute of Aerodynamics and Gas Dynamics, University of Stuttgart
Lecturers: Dr. Volker Weinberg (LRZ) / Dr. Momme Allalen (LRZ)
AGENDA
09:00 - 10:30 Guest lecture about Deep Neural Networks for Data-Driven Turbulence Models, using DL in CFD
10:30-10:45 Coffee Break
10:45-11:00 Intro
11:00-12:00 Profiling
12:00-13:00 Lunch Break
13:00-13:30 Introduction into NVIDIA® Nsight™ Systems
13:30-15:00 OpenACC Directives
15:00-15:30 Coffee Break
15:30-16:15 GPU Programming
16:15-16:45 Data Management and Loop Optimizations
16:45-17:00 Q&A, Final Remarks
SLIDES
DOCUMENTATION
4th day: ML Examples and Methods on HLRS systems (9:00 - 16:30)
Presenters: Dr. Lorenzo Zanon, Patrick Vogler, Dr. Khatuna Kakhiani (HLRS)
On the 4th day we cover Pre-processing of data, mathematical methods and machine learning as well as Deep learning on HLRS Systems.
AGENDA
09:00-12:30 Focus on Pre-processing, Feature Engineering and Machine Learning: Stuttgart S-Bahn Example (Dr. Lorenzo Zanon - presenter, Li Zhong and Dennis Hoppe, HLRS)
12:30-13:30 Lunch Break
13:30-14:30 Data Compression of numerical data sets with the BigWhoop library (Patrick Vogler, HLRS in cooperation with EXCELLERAT)
14:30-16:30 Waste classification using Deep learning (Dr. Khatuna Kakhiani, HLRS)
LOGIN REQUIREMENTS
Access with testuser / password: sbahn2020
https://fs.hlrs.de/people/zanon/Vulcan_login_requirements.pdf
https://fs.hlrs.de/people/zanon/Enkidu_login_requirements.pdf
https://fs.hlrs.de/people/zanon/07_01_requirements.pdf (key generation and Firefox profile for JN)
SLIDES
DeepLearningWorkshop2020-BigWhoop.7z
read_data_12.07_EDITS_kakhiani.ipynb
Survey
- Please fill out the survey under https://events.prace-ri.eu/event/1040/surveys/680
- This helps us
- increase the quality of the courses,
- design the future training programme at HLRS and LRZ and in Europe according to your needs and wishes,
- get future funding for training events.
Information on ZOOM
- ZOOM help center has great resources with help articles and videos for getting started: https://support.zoom.us/hc/en-us
- This “Getting Started” page is a great resource: https://support.zoom.us/hc/en-us/categories/200101697
- This FAQ has tons of useful info: https://support.zoom.us/hc/en-us/articles/206175806-Frequently-Asked-Questions
- We strongly encourage you to read some of the basic info relevant to your operating system:
● Getting Started on Windows and Mac: https://support.zoom.us/hc/en-us/articles/201362033-Getting-Started-on-Windows-and-Mac
● Getting Started On Chrome OS: https://support.zoom.us/hc/en-us/articles/213298746-Getting-Started-On-Chrome-OS - You may log in via the app or a browser. We recommend downloading the app for the best experience. It may take several minutes to download, so if you are using Zoom for the first time, please download the app prior to the event, https://zoom.us/download , or join the meeting early.
- You may sign up for a free account at zoom.us/signup . Or, you may join a meeting as a guest without a Zoom account.
NEXT STEPS
Visit the NVIDIA Deep Learning lnstitute's website at http://www.nvidia.co.uk/dli to access more training and resources.
- Start online, self-paced training in deep learning and accelerated computing (using the account you created today).
- View upcoming workshops around the world and request an onsite workshop at your company or organization.
- Learn about the University Ambassador Program.
Ready to kick off a deep learning project or already working on one? Choose the best software and hardware solutions at
http://www.nvidia.co.uk/deep-learning-ai/developer/
