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:20220812T074334Z
LOCATION:Sydney Room
DTSTART;TZID=Europe/Stockholm:20220627T143000
DTEND;TZID=Europe/Stockholm:20220627T150000
UID:submissions.pasc-conference.org_PASC22_sess117_msa215@linklings.com
SUMMARY:Online Machine Learning for ESM: Using PyTorch and TensorFlow in F
 ortran with SmartSim
DESCRIPTION:Minisymposium\n\nOnline Machine Learning for ESM: Using PyTorc
 h and TensorFlow in Fortran with SmartSim\n\nRigazzi\n\nThe inclusion of M
 achine Learning (ML) into Earth System Models (ESM) presents unique domain
  science and engineering challenges. SmartSim is an open source library de
 dicated to addressing the engineering challenges around integrating ML int
 o ESM, and other HPC simulations like ESMs, at scale. SmartSim provides th
 e ability to call popular ML frameworks like TensorFlow and PyTorch from w
 ithin HPC simulations written in Fortran, C, C++, and Python. <br /><br />
 In this talk, we demonstrate how SmartSim was used to integrate a PyTorch 
 based eddy-kinetic energy parameterization within a Fortran based ocean mo
 del, Modular Ocean Model 6 (MOM6). We will show the results of the ML inte
 gration, scaling details, and provide implementation instructions for user
 s who wish to try themselves.\n\nDomain: Climate, Weather and Earth Scienc
 es, Computer Science and Applied Mathematics
END:VEVENT
END:VCALENDAR