projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 891605b) | patch
[ncnn-backend] Support add operation (#8565)
author오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 <hseok82.oh@samsung.com>
Tue, 29 Oct 2019 11:31:27 +0000 (20:31 +0900)
committer이한종/On-Device Lab(SR)/Engineer/삼성전자 <hanjoung.lee@samsung.com>
Tue, 29 Oct 2019 11:31:27 +0000 (20:31 +0900)
commit7c8f88cb77a0b37a67bc796a8b9f99dbfdbb2ca9
tree86e7b0f8e56157b3760a87d531b20b83cdab1166tree | snapshot
parent891605b712b512ab66745cf61d03d8324888a771commit | diff
[ncnn-backend] Support add operation (#8565)

- NCNN binary-op kernel porting (disable same input/output tensor support)
- NCNN backend add operation support
  - Support non-broadcasting
  - Need alignment matching when rank is upper 3
- Enable unittest, related validation test

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
compute/ncnn/include/ncnn/layer/binaryop.h [new file with mode: 0644] blob
compute/ncnn/src/layer/arm/neon_mathfun.h [new file with mode: 0644] blob
compute/ncnn/src/layer/binaryop.cc [new file with mode: 0644] blob
runtimes/neurun/backend/srcn/KernelGenerator.cc diff | blob | history
runtimes/neurun/backend/srcn/KernelGenerator.h diff | blob | history
runtimes/neurun/backend/srcn/ShapeFixer.cc diff | blob | history
runtimes/neurun/backend/srcn/ShapeFixer.h diff | blob | history
runtimes/neurun/backend/srcn/kernel/AddLayer.cc [new file with mode: 0644] blob
runtimes/neurun/backend/srcn/kernel/AddLayer.h [new file with mode: 0644] blob
tests/nnapi/nnapi_gtest.skip.armv7l-linux.ncnn diff | blob | history
Domain: Machine Learning / ML Framework;