projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 29443ed) | patch
[neurun] Implement PermuteLayer kernel for cpu (#2749)
author이한종/동작제어Lab(SR)/Engineer/삼성전자 <hanjoung.lee@samsung.com>
Tue, 18 Sep 2018 10:57:04 +0000 (19:57 +0900)
committer이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 <chunseok.lee@samsung.com>
Tue, 18 Sep 2018 10:57:04 +0000 (19:57 +0900)
commita04439ab3aae69dae74cf1c03caf225dd8c13471
tree48063e878b529a5d8883148ce77d113a49b8b95dtree | snapshot
parent29443edf6833ed78165ac4772ea3721f8d0e176dcommit | diff
[neurun] Implement PermuteLayer kernel for cpu (#2749)

PermuteLayer basically copies data from input tensor to output tensor,
but if the tensor must be permuted it also do permute the tensor.
Currently NHWC to NCHW and NCHW to NHWC permutation is supported.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
runtimes/neurun/src/kernel/cpu/PermuteLayer.cc [new file with mode: 0644] blob
runtimes/neurun/src/kernel/cpu/PermuteLayer.h [new file with mode: 0644] blob
Domain: Machine Learning / ML Framework;