https://github.com/Microsoft/CNTK
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
CompositeDataReader.cpp -rw-r--r-- 12.6 KB
CompositeDataReader.h -rw-r--r-- 3.9 KB
CompositeDataReader.vcxproj -rw-r--r-- 4.4 KB
CompositeDataReader.vcxproj.filters -rw-r--r-- 497 bytes
Exports.cpp -rw-r--r-- 1.4 KB
dllmain.cpp -rw-r--r-- 434 bytes
stdafx.cpp -rw-r--r-- 402 bytes
stdafx.h -rw-r--r-- 695 bytes
targetver.h -rw-r--r-- 466 bytes

back to top