projects / platform / core / ml / nnfw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8868d71) | patch
[PureCL] Support TENSOR_QUANT8_ASYMM type operand (#1623)
author서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 <sangmin7.seo@samsung.com>
Fri, 8 Jun 2018 08:06:56 +0000 (17:06 +0900)
committerGitHub Enterprise <noreply-CODE@samsung.com>
Fri, 8 Jun 2018 08:06:56 +0000 (17:06 +0900)
commitab56911301c425fd5b5912249c844410fe7bcd53
treee8edd43acf5d83e298ce939909f159e103a19873tree | snapshot
parent8868d715ac475a7d5c2709cef9578bc4e167e351commit | diff
[PureCL] Support TENSOR_QUANT8_ASYMM type operand (#1623)

For issue #1481

This patch revises `internal::tflite::operand::Object` to keep `scale`
and `zeroPoint` in order to support operand of
ANEURALNETWORKS_TENSOR_QUANT8_ASYMM type, and then modifies
`ANeuralNetworksModel_addOperand` to pass the scale and zeroPoint values
as well when appending a new operand.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
runtimes/pure_arm_compute/src/internal/Model.cc diff | blob | history
runtimes/pure_arm_compute/src/internal/Model.h diff | blob | history
runtimes/pure_arm_compute/src/model.cc diff | blob | history
Domain: Machine Learning / ML Framework;