swh:1:snp:f50ab94432af916b5fb8b4ad831e8dddded77084
History
Tip revision: 5cc66af4ddf6c2ed9574e3e8d7b2ba895cb8da58 authored by Amit Agarwal on 13 December 2016, 06:34:50 UTC
CNTK parallel training: Turned off quantization for gradient matrices where row count is too small to benefit from quantization
Tip revision: 5cc66af
File Mode Size
Common
Config
Control
Data
AccumulatorNodeTests.cpp -rw-r--r-- 6.1 KB
CropNodeTests.cpp -rw-r--r-- 9.8 KB
NetworkTests.vcxproj -rw-r--r-- 6.2 KB
NetworkTests.vcxproj.filters -rw-r--r-- 1.9 KB
OperatorEvaluation.cpp -rw-r--r-- 837 bytes
TestHelpers.cpp -rw-r--r-- 2.6 KB
TestHelpers.h -rw-r--r-- 1.3 KB
stdafx.cpp -rw-r--r-- 400 bytes
stdafx.h -rw-r--r-- 593 bytes
targetver.h -rw-r--r-- 465 bytes

back to top