~/talk

PyTorch at Microsoft

on YouTube 07m
Talkat PyTorch Developer Conference 2019 · 2019-10-08..10-11· [machine-learning]

speaker

name:Saurabh Tiwary
links:none on file

resources

video:https://youtu.be/_NABaLrrZkE
slides:none on file

summary

PyTorch becomes the production DL stack for training at scale with native Windows support, ONNX for runtime acceleration, and model/data parallelism to handle huge nets. Teams train a 1.2 billion parameter model (Yule) that covers 100+ languages with weaker results on low‑resource ones, then compress with knowledge distillation and ONNX to deploy across CPUs, GPUs and FPGAs. A deployed pipeline hits about 1,000,000 inferences per second on FPGA for a 30–40M model and powers Bing, Outlook, Powerpoint Designer.

AI-written summary · about

chapters (6)

  1. 0:00Intro
  2. 0:56Internal adoption
  3. 1:29Internal tooling
  4. 2:44Model training
  5. 4:08Inference
  6. 5:35Applications

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.