This is an automated email from the ASF dual-hosted git repository.
wangwei pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-singa.git.
from 7530b44 Merge pull request #533 from dcslin/add-fix-tensor-assert
new 03687eb Add Sync() function in CudaGpu Device to assist time profiling
new f6d5f2d SINGA-490 Optimize performance of stochastic gradient descent
(SGD)
new a8db8fa SINGA-490 Replace with in-place elementwise mult
new ae25e1c Add the Sync() in the SGD loop
new 0080369 Merge pull request #535 from chrishkchris/SINGA-490
The 1397 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
examples/autograd/resnet.py | 14 +++++++++-----
examples/autograd/resnet_dist.py | 23 ++++++++++++++---------
include/singa/core/device.h | 3 ++-
python/singa/autograd.py | 38 +++++++++++++++++++++-----------------
python/singa/opt.py | 10 +++++-----
src/api/core_device.i | 1 +
src/core/device/cuda_gpu.cc | 4 ++++
src/core/tensor/tensor_math_cuda.h | 8 ++------
8 files changed, 58 insertions(+), 43 deletions(-)
