~/talk

Leverage AI Kits for Faster AI Solutions Tailored to Your Needs

on YouTube 22m
Talkat Microsoft Build 2023 · 2023-05-23..05-25· [machine-learning, ai-assisted-development]

speakers

name:Rob DeRider
links:none on file
name:Bereket Gaguro
links:none on file
name:Michelle Megarry
links:none on file
name:Brian Rogers
links:none on file
name:Kiana Yup
links:none on file

resources

video:https://youtu.be/pZGAdh2xr9U
slides:none on file

summary

Intel provides 34 reference kits using oneAPI toolkits (Modin, IPEX, OpenVINO, Intel Neural Compressor) plus SigOpt for HPO and GitHub repos; a Scikit‑Learn SVC doc‑indexing kit swaps in Modin and sklearn‑extension to deliver 2.5X boost in data processing and 25X training speed with no accuracy loss. A ResNet‑50 demo shows stock PyTorch ~107 images/sec vs an int8 quantized IPEX model ~1600 images/sec (~15X), running on Sapphire Rapids AMX.

AI-written summary · about

chapters (5)

  1. 0:00Introduction
  2. 0:34AI Reference Kits
  3. 6:38Intel Tools for Developers
  4. 11:09Demo
  5. 17:25Why Quantize

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.