BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082409Z
LOCATION:HG E 3
DTSTART;TZID=Europe/Stockholm:20240604T170000
DTEND;TZID=Europe/Stockholm:20240604T173000
UID:submissions.pasc-conference.org_PASC24_sess139_msa129@linklings.com
SUMMARY:TorchFort: A Library for Online Deep Learning in Fortran HPC Progr
 ams
DESCRIPTION:Minisymposium\n\nThorsten Kurth, Josh Romero, and Massimiliano
  Fatica (NVIDIA Inc.)\n\nDeep learning has shown promise in reducing compu
 tational cost or as an alternative method for modeling physical phenomena 
 for a broad range of scientific applications. In these domains, the data s
 ources are numerical simulation programs typically implemented in C, C++, 
 or still often, Fortran. This is in contrast to popular deep learning fram
 eworks that users interact with using Python. A source of friction that of
 ten arises is how to efficiently couple the simulation program with the DL
  framework for training or inference. \n\nIn this talk, we discuss TorchFo
 rt, a library for online DL training and inference implemented with LibTor
 ch, the C++ backend used by PyTorch. This library can be invoked directly 
 from Fortran/C/C++, enabling transparent sharing of data arrays from the s
 imulation program to the DL framework, all contained within the simulation
  process. We will talk about the library design and some implementation ex
 amples to present opportunities this tight coupling presents for DL applic
 ations.\n\nDomain: Chemistry and Materials, Climate, Weather, and Earth Sc
 iences, Engineering, Computational Methods and Applied Mathematics\n\nSess
 ion Chair: Alessandro Rigazzi (HPE)
END:VEVENT
END:VCALENDAR