projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e54b4f0) | patch
[optimizer] Move optimizer variables to weights
authorParichay Kapoor <pk.kapoor@samsung.com>
Thu, 3 Dec 2020 05:43:09 +0000 (14:43 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Thu, 10 Dec 2020 10:20:41 +0000 (19:20 +0900)
commita023b3ace5a999c4841351bdf717b8bf8f7943cc
tree9e3f49b4acd5ff41b39f0ea6f895806cdcad7b27tree | snapshot
parente54b4f0903152d14c6261370fd43ca22be150f05commit | diff
[optimizer] Move optimizer variables to weights

Move optimizer variables to weights
Now all the weight related tensors are handled by weights themselves
So, optimizer can be shared across all layers, no need to create new
copies for all layers

**Self evaluation:**
1. Build test: [x]Passed [ ]Failed [ ]Skipped
2. Run test: [x]Passed [ ]Failed [ ]Skipped

Signed-off-by: Parichay Kapoor <pk.kapoor@samsung.com>
nntrainer/layers/layer.cpp diff | blob | history
nntrainer/optimizers/adam.cpp diff | blob | history
nntrainer/optimizers/adam.h diff | blob | history
nntrainer/tensor/weight.h diff | blob | history
Domain: Machine Learning / ML Framework;