projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b89c9f0) | patch
[enco] Use Conv2D kernel data without modification (#1858)
author박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 <jh1302.park@samsung.com>
Mon, 15 Oct 2018 02:22:58 +0000 (11:22 +0900)
committerGitHub Enterprise <noreply-CODE@samsung.com>
Mon, 15 Oct 2018 02:22:58 +0000 (11:22 +0900)
commitd293fab136f39e83a11a8e81dafee5dd4325d85e
treedd938d91049021361252967b330d1bf6c0df21f7tree | snapshot
parentb89c9f0aabcf36c8ad1744aff134f16e9033f978commit | diff
[enco] Use Conv2D kernel data without modification (#1858)

This commit revises Split pass to use Conv2D kernel data that coco::Data
provides without modification (to reduce peak memory consumption).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
contrib/enco/core/src/Transforms/Split.cpp diff | blob | history
Domain: Machine Learning / ML Framework;