projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e11d175) | patch
[loss] BugFix for regularization loss
authorParichay Kapoor <pk.kapoor@samsung.com>
Mon, 2 Aug 2021 08:55:05 +0000 (17:55 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Tue, 3 Aug 2021 02:07:46 +0000 (11:07 +0900)
commitb9abb5981e1e5e4c8fc06bfb363440b78c744068
tree0870fab768d43007d5e8fc4e0d5bb6eb68865071tree | snapshot
parente11d175bba83a95107db631cfb90538b60246251commit | diff
[loss] BugFix for regularization loss

This patch adds bugfix for regularization loss calculation.
Regularization loss must be calculated before the weight is updated in
backwarding, but this was not being done.
This patch provides the corresponding fix.

Signed-off-by: Parichay Kapoor <pk.kapoor@samsung.com>
nntrainer/layers/common_properties.h diff | blob | history
nntrainer/layers/layer_context.h diff | blob | history
nntrainer/layers/layer_node.cpp diff | blob | history
nntrainer/layers/layer_node.h diff | blob | history
test/unittest/unittest_base_properties.cpp diff | blob | history
Domain: Machine Learning / ML Framework;