BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T100915Z
UID:c4d55c4d-93d9-4b16-83b5-c345109dc097
DTSTART:20200907T070000Z
DTEND:20200910T130000Z
DESCRIPTION:\nThis workshop will be delivered as an ONLINE COURSE for remot
 e participation \ndue to the COVID-19 measures enforced by most European 
 governments.\n\nThe workshop will take place ONLINE via Zoom on MONDAY\, 7
  - THURSDAY\, 10 SEPTEMBER 2020 at 10:00-12:00 and 13:00-16:00 EEST [09:00
 -11:00 and 12:00-15:00 CEST] each day.\n\n\n\nOverview\n\nNVIDIA Deep Lear
 ning Institute (DLI) offers hands-on training for developers\, data scien
 tists\, and researchers looking to solve challenging problems with deep le
 arning.\n\nLearn how to train and deploy a neural network to solve real-wo
 rld problems\, how to generate effective descriptions of content within im
 ages and video clips\, how to effectively parallelize training of deep neu
 ral networks on Multi-GPUs and how to accelerate your applications with CU
 DA C/C++ and OpenACC.\n\nThis 4-days workshop  combines lectures about fu
 ndamentals of Deep Learning for Multiple Data Types and Multi-GPUs with le
 ctures about Accelerated Computing with CUDA C/C++ and OpenACC.\n\nThe lec
 tures are interleaved with many hands-on sessions using Jupyter Notebooks.
  The exercises will be done on a fully configured GPU-accelerated workstat
 ion in the cloud.\n\nThe workshop is part of PRACE Training Centres activi
 ty and co-organized by LRZ – Leibniz Supercomputing Centre (Garching nea
 r Munich) as part of Gauss Centre for Supercomputing (Germany)\, IT4I – 
 National Supercomputing Center VSB Technical University of Ostrava (Czech 
 Republic)\, CSC – IT Center for Science Ltd (Finland) and NVIDIA Deep Le
 arning Institute (DLI) for the Partnership for Advanced Computing in Europ
 e (PRACE).\n\nLecturers:  Dr. Momme Allalen\, Dr. Juan Durillo Barrionuev
 o\, Dr. Volker Weinberg (LRZ and NVIDIA University Ambassadors)\, Georg Zi
 tzlsberger (IT4Innovations and NVIDIA University Ambassador)\n\nLanguage:
   English\nPrice:           Free of charge (4 training days)\n\
 nPrerequisites and content level\n\nPlease note\, that the workshop is exc
 lusively for verifiable students\, staff\, and researchers from any academ
 ic institution (for industrial participants\, please contact NVIDIA for in
 dustrial specific training).\n\nTechnical background\, basic understanding
  of machine learning concepts\, basic C/C++ or Fortran programming skills.
  In addition\, basics in Python will be helpful. Since Python 2.7 is used\
 , the following tutorial can be used to learn the syntax: docs.python.org/
 2.7/tutorial/index.html.\n\nFor the 1st day familiarity with TensorFlow wi
 ll be a plus as all the hands-on sessions are using TensorFlow. For those 
 who do not program in TensorFlow\, please go over TensorFlow tutorial (esp
 ecially the "Learn and use ML" section): www.tensorflow.org/tutorials/.\n\
 nThe content level of the course is broken down as: beginner's - 5\,2 h (2
 0%)\, intermediate - 14\,3 h (55%)\, advanced - 6\,5 h (25%)\, community-t
 argeted content - 0\,0 h (0%).\n\nImportant information\n\nAfter you are a
 ccepted\, please create an account under courses.nvidia.com/join .\n\nEnsu
 re your laptop / PC will run smoothly by going to http://websocketstest.co
 m/\n\nMake sure that WebSockets work for you by seeing under Environment\,
  WebSockets is supported and Data Receive\, Send and Echo Test all check Y
 es under WebSockets (Port 80). If there are issues with WebSockets\, try u
 pdating your browser. If you have any questions\, please contact Marjut Di
 eringer at: mdieringer"at"nvidia.com. \n\nAGENDA / Description and learni
 ng outcomes\n\nDay 1: Fundamentals of Deep Learning for Multiple Data Type
 s\n\nThis day explores how convolutional and recurrent neural networks can
  be combined to generate effective descriptions of content within images a
 nd video clips.\n\nLearn how to train a network using TensorFlow and the M
 icrosoft Common Objects in Context (COCO) dataset to generate captions fro
 m images and video by:\n\n\n	Implementing deep learning workflows like ima
 ge segmentation and text generation\n	Comparing and contrasting data types
 \, workflows\, and frameworks\n	Combining computer vision and natural lang
 uage processing\n\n\nUpon completion\, you’ll be able to solve deep lear
 ning problems that require multiple types of data inputs.\n\nDay 2: Fundam
 entals of Accelerated Computing with OpenACC\n\nOn the 2d day you learn th
 e basics of OpenACC\, a high-level programming language for programming on
  GPUs. Discover how to accelerate the performance of your applications bey
 ond the limits of CPU-only programming with simple pragmas. You’ll learn
 :\n\n\n	How to profile and optimize your CPU-only applications to identify
  hot spots for acceleration\n	How to use OpenACC directives to GPU acceler
 ate your codebase\n	How to optimize data movement between the CPU and GPU 
 accelerator\n\n\nUpon completion\, you'll be ready to use OpenACC to GPU a
 ccelerate CPU-only applications.\n\nDay 3: Fundamentals of Accelerated Com
 puting with CUDA C/C++\n\nThe CUDA computing platform enables the accelera
 tion of CPU-only applications to run on the world’s fastest massively pa
 rallel GPUs. On the 3rd day you experience C/C++ application acceleration 
 by:\n\n\n	Accelerating CPU-only applications to run their latent paralleli
 sm on GPUs\n	Utilizing essential CUDA memory management techniques to opti
 mize accelerated applications\n	Exposing accelerated application potential
  for concurrency and exploiting it with CUDA streams\n	Leveraging command 
 line and visual profiling to guide and check your work\n\n\nUpon completio
 n\, you’ll be able to accelerate and optimize existing C/C++ CPU-only ap
 plications using the most essential CUDA tools and techniques. You’ll un
 derstand an iterative style of CUDA development that will allow you to shi
 p accelerated applications fast.\n\nDay 4: Fundamentals of Deep Learning f
 or Multi-GPUs\n\nThe computational requirements of deep neural networks us
 ed to enable AI applications like self-driving cars are enormous. A single
  training cycle can take weeks on a single GPU or even years for larger da
 tasets like those used in self-driving car research. Using multiple GPUs f
 or deep learning can significantly shorten the time required to train lots
  of data\, making solving complex problems with deep learning feasible.\n\
 nOn the last day we will teach you how to use multiple GPUs to train neura
 l networks. You'll learn:\n\n\n	Approaches to multi-GPUs training\n	Algori
 thmic and engineering challenges to large-scale training\n	Key techniques 
 used to overcome the challenges mentioned above\n\n\nUpon completion\, you
 'll be able to effectively parallelize training of deep neural networks us
 ing TensorFlow.\nhttps://events.prace-ri.eu/event/998/
SUMMARY:[ONLINE] Deep Learning and GPU Programming Workshop @ CSC
URL;VALUE=URI:https://events.prace-ri.eu/event/998/
END:VEVENT
END:VCALENDAR