projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 0fdbb43) | patch
[rnn] Cleanup RNN implementation
authorParichay Kapoor <kparichay@gmail.com>
Thu, 8 Jul 2021 07:26:04 +0000 (16:26 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Thu, 22 Jul 2021 11:47:24 +0000 (20:47 +0900)
commit7015d84c4514cb753d0ff8106fb7fc6b6aaaedb6
treea52e1e5077ca66d247f34779b01587099b49a6d4tree | snapshot
parent0fdbb43cc803a3c7989a12dfcd9af1f5a27fe301commit | diff
[rnn] Cleanup RNN implementation

Cleanup RNN implementation.
Reduce usage of temporary allocated memory and reuse existing memory as much as possible.
Further, request hidden state memory from context than creating by self.

Signed-off-by: Parichay Kapoor <kparichay@gmail.com>
nntrainer/layers/rnn.cpp diff | blob | history
nntrainer/layers/rnn.h diff | blob | history
test/unittest/layers/unittest_layers_rnn.cpp diff | blob | history
Domain: Machine Learning / ML Framework;