projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d135a82) | patch
renamed previous mul to matmul (#1254)
author윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 <hyunsik.yoon@samsung.com>
Thu, 17 May 2018 23:31:23 +0000 (08:31 +0900)
committer박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 <saehie.park@samsung.com>
Thu, 17 May 2018 23:31:23 +0000 (08:31 +0900)
commit19d5b52de1fd744d1994d142f4a4e413cdb9249b
treeb3f69de268c98dede199fccdd90a3346823fec11tree | snapshot
parentd135a822db72ea9a6529f1edd049bd8b1d0bb984commit | diff
renamed previous mul to matmul (#1254)

In previous PR (#1238), `mul` was added for two matrix multiplication.
However, it turned out that `mul` must be element-wise mul, not matmul (matrix mul).
For this reason previous `mul` was renamed to `matmul` and test was changed to `disabled`.
Also, 'matmul2x2.tflite` was uploaded into `npuarchive.mooo.com`.
(previous `mul2x2.tflite` in `npuarchive.mooo.com` will be deleted after this PR is merged.)

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
tests/framework/tests/mul/matmul2x2/config.sh [new file with mode: 0644] blob
tests/framework/tests/mul/mul2x2/config.sh [deleted file] blob | history
tools/tensorflow_model_freezer/sample/matmul2x2_freezer.py [moved from tools/tensorflow_model_freezer/sample/mul2x2_freezer.py with 79% similarity] diff | blob | history
tools/tensorflow_model_freezer/sample/run_converter.py diff | blob | history
Domain: Machine Learning / ML Framework;