projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: dbe2eb1) | patch
Apply optimized cpu kernel of MaxPoolFloat32 (#7886)
author장지섭/On-Device Lab(SR)/Engineer/삼성전자 <jiseob.jang@samsung.com>
Wed, 2 Oct 2019 05:35:42 +0000 (14:35 +0900)
committer이춘석/On-Device Lab(SR)/Staff Engineer/삼성전자 <chunseok.lee@samsung.com>
Wed, 2 Oct 2019 05:35:42 +0000 (14:35 +0900)
commitfdf8a841bdcfb167e53943e91ab758a8c8f437e3
tree1038590b3523948ef638753017f8ff205f0f6c2ctree | snapshot
parentdbe2eb10a18c3ac8e9a2419362c887e95e4813c1commit | diff
Apply optimized cpu kernel of MaxPoolFloat32 (#7886)

This commit applies optimized cpu kernel of MaxPoolFloat32.
  - Introduce optimized cpu kernel of MaxPool for float
  - Apply that kernel for cpu backend of neurun

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
runtimes/libs/cker/include/cker/Types.h diff | blob | history
runtimes/libs/cker/include/cker/operation/AveragePool.h diff | blob | history
runtimes/libs/cker/include/cker/operation/MaxPool.h diff | blob | history
runtimes/libs/cker/include/cker/operation/optimized/AveragePool.h diff | blob | history
runtimes/libs/cker/include/cker/operation/optimized/MaxPool.h [new file with mode: 0644] blob
runtimes/libs/cker/include/cker/operation/reference/AveragePool.h diff | blob | history
runtimes/libs/cker/include/cker/operation/reference/MaxPool.h [new file with mode: 0644] blob
runtimes/neurun/backend/cpu/kernel/AvgPoolLayer.cc diff | blob | history
runtimes/neurun/backend/cpu/kernel/MaxPoolLayer.cc diff | blob | history
runtimes/neurun/core/src/exec/interp/operations/AvgPool2D.cc diff | blob | history
runtimes/neurun/core/src/exec/interp/operations/MaxPool2D.cc diff | blob | history
Domain: Machine Learning / ML Framework;