[Torch, QNN] Add support for quantized models via QNN - #4977

Merged

anijain2305 merged 14 commits into

apache:masterfrom

masahi:torch-qnn

Mar 4, 2020

Commits on Mar 1, 2020

qnn support initial import
masahi
committed
29e8644

Commits on Mar 2, 2020

fix upsampling num input
masahi
committed
e7df8ec
imagenet tests added
masahi
committed
f5a319f
add qunatized module tests
masahi
committed
93f374a
quantized module tests working
masahi
committed
0615ca6
imagenet test working
masahi
committed
da89492
fix lint
masahi
committed
0daf4ca
remove top level torch import to fix ci error
masahi
committed
6a99ec9
disable lint warning on outside toplevel import
masahi
committed
96ccb37
revert parse -> convert change
masahi
committed
75c1200
add comments to qnn translation
masahi
committed
9c9556d
address comments, add sample outputs
masahi
committed
a6239e1
add more comments
masahi
committed
d33ac99

Commits on Mar 3, 2020

refactor bias add and requantize step
masahi
committed
80b1a61