projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 4f97a14) | patch
[neurun] Introduce tensor conversion Nodes (#2064)
author김수진/동작제어Lab(SR)/Engineer/삼성전자 <sjsujin.kim@samsung.com>
Thu, 26 Jul 2018 02:04:36 +0000 (11:04 +0900)
committer이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 <chunseok.lee@samsung.com>
Thu, 26 Jul 2018 02:04:36 +0000 (11:04 +0900)
commit5a54b9e30770c23126c44b10769abb97f442d89f
treee5829e79554822c508b21e86cd89663c06cf974dtree | snapshot
parent4f97a145a010ef7e8dee1cbcba523f7fd1192947commit | diff
[neurun] Introduce tensor conversion Nodes (#2064)

* [neurun] Introduce tensor conversion Nodes

Related : #1974

This commit introduces tensor conversion Nodes.

- FromCommon : "common order shape" -> " specific shape of next op".
- ToCommon : "specific shape of prev op" -> "common shape".

The name of 'Common' would be changed to another.

It's part PR of #2056(https://github.sec.samsung.net/STAR/nnfw/pull/2056#issuecomment-128722)

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Remove unnecessary op list

* Seperate ACL/CPU converter Nodes
runtimes/neurun/src/backend/acl_cl/StageGenerator.cc diff | blob | history
runtimes/neurun/src/backend/acl_cl/StageGenerator.h diff | blob | history
runtimes/neurun/src/backend/cpu/StageGenerator.cc diff | blob | history
runtimes/neurun/src/backend/cpu/StageGenerator.h diff | blob | history
runtimes/neurun/src/compilation.cc diff | blob | history
runtimes/neurun/src/internal/IStageGenerator.h diff | blob | history
runtimes/neurun/src/internal/op/NodeVisitor.h diff | blob | history
runtimes/neurun/src/internal/op/TensorConvert_Acl.cc [new file with mode: 0644] blob
runtimes/neurun/src/internal/op/TensorConvert_Acl.h [new file with mode: 0644] blob
runtimes/neurun/src/internal/op/TensorConvert_Cpu.cc [new file with mode: 0644] blob
runtimes/neurun/src/internal/op/TensorConvert_Cpu.h [new file with mode: 0644] blob
Domain: Machine Learning / ML Framework;