HomePhabricator
Diffusion LLVM Github Monorepo 27d776fa6d04

Convert per channel fake quant attributes to type
27d776fa6d04
Actions

Authored by Feng Liu <fengliuai@google.com> on Sep 9 2019, 2:57 PM.

Description

Convert per channel fake quant attributes to type

For per channel fake quant attributes, the returned type should be
UniformQuantizedPerAxisType. Currently, this method isn't under test because we
haven't added the quant_ConstFakeQuantPerAxis op and the convert method.

PiperOrigin-RevId: 268084017

Details

Committed
A. Unique TensorFlower <gardener@tensorflow.org>Sep 9 2019, 2:57 PM
Parents
rG893c86fff7ef: Explicitly declare the OpPassManager move constructor to avoid undefined errors.
Branches
Unknown
Tags
Unknown

Event Timeline

A. Unique TensorFlower <gardener@tensorflow.org> committed rG27d776fa6d04: Convert per channel fake quant attributes to type (authored by liufengdb).Sep 9 2019, 2:57 PM

Changes (2)

PathSize
mlir/
include/
mlir/
Dialect/
QuantOps/
lib/
Dialect/
QuantOps/
Utils/

rG27d776fa6d04

View Options

mlir/include/mlir/Dialect/QuantOps/FakeQuantSupport.h

Loading...
View Options

mlir/lib/Dialect/QuantOps/Utils/FakeQuantSupport.cpp

Loading...