BEGIN:VCALENDAR PRODID:-//Microsoft Corporation//Outlook 16.0 MIMEDIR//EN VERSION:2.0 METHOD:PUBLISH X-MS-OLK-FORCEINSPECTOROPEN:TRUE BEGIN:VTIMEZONE TZID:Pacific Standard Time BEGIN:STANDARD DTSTART:16011104T020000 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 END:STANDARD BEGIN:DAYLIGHT DTSTART:16010311T020000 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT CLASS:PUBLIC CREATED:20191016T235426Z DESCRIPTION:Maggie Zhang\, Nathan Luehr\, Josh Romero\, Pooya Davoodi\, and Davide Onofrio dive into techniques to accelerate deep learning training and inference for common deep learning and machine learning workloads. You ’ll learn how DALI can eliminate input/output (I/O) and data processing bottlenecks in real-world applications and how automatic mixed precision ( AMP) can easily give you up to 3x training performance improvement on Volt a GPUs. You’ll see best practices for multi-GPU and multinode scaling us ing Horovod. They use a deep learning profiler to visualize the TensorFlow operations and identify optimization opportunities. And you’ll learn to deploy these trained models using INT8 quantization in TensorRT (TRT)\, a ll within new convenient APIs of the TensorFlow framework.\n \nWhat you'll learn\n* Discover components from NVIDIA’s software stack to speed up p ipelines and eliminate I/O bottlenecks\n* Learn how to enable mixed precis ion when training models and use TRT to optimize your trained models for i nference\n DTEND;TZID="Pacific Standard Time":20191029T170000 DTSTAMP:20191016T235426Z DTSTART;TZID="Pacific Standard Time":20191029T133000 LAST-MODIFIED:20191016T235426Z LOCATION:TensorFlow World 2019: Grand Ballroom E PRIORITY:5 SEQUENCE:0 SUMMARY;LANGUAGE=en-us:NVIDIA Hands-On Tutorial: Accelerating training\, in ference\, and ML applications on NVIDIA GPUs TRANSP:OPAQUE UID:040000008200E00074C5B7101A82E00800000000D0E5CCF34184D501000000000000000 010000000565750C802806842A7764F1283BBD700 X-ALT-DESC;FMTTYPE=text/html:\n\n\n

Maggie Zhang\, Nathan Luehr\, Josh Ro mero\, Pooya Davoodi\, and Davide Onofrio dive into techniques to accelera te deep learning training and inference for common deep learning and machi ne learning workloads. You’\;ll learn how \;DALI \;can eliminate input/output (I/O) an d data processing bottlenecks in real-world applications and how automatic mixed precision (AMP) ca n easily give you up to 3x training performance improvement on Volta GPUs. You’\;ll see best practices for multi-GPU \;and multinode scali ng using Horovod. They use a deep learning profi ler to visualize the TensorFlow operations and identify optimization oppor tunities. And you’\;ll learn to deploy these trained models using INT 8 quantization in TensorRT (TRT)\, all within new convenient APIs of the TensorFlow framework.

 \;

What you'll learn

X-MICROSOFT-CDO-BUSYSTATUS:BUSY X-MICROSOFT-CDO-IMPORTANCE:1 X-MICROSOFT-DISALLOW-COUNTER:FALSE X-MS-OLK-AUTOFILLLOCATION:FALSE X-MS-OLK-CONFTYPE:0 BEGIN:VALARM TRIGGER:-PT15M ACTION:DISPLAY DESCRIPTION:Reminder END:VALARM END:VEVENT END:VCALENDAR