platform/core/ml/nnfw.git
6 years agoAdd failed test(preview4) into skiplist (#1934)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Jul 2018 01:52:12 +0000 (10:52 +0900)]
Add failed test(preview4) into skiplist (#1934)

For generated tests of android-p-preview4, add failed test into skiplist

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoUpdate ARM CL to v18.05 (#1925)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Jul 2018 01:51:33 +0000 (10:51 +0900)]
Update ARM CL to v18.05 (#1925)

Patch to migrate ARM CL v18.05
- Use arm_compute/graph/frontend API to build graph data structure
- Update ACL submodule: v18.05

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoAdd explicit padding to conv (#1936)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Jul 2018 01:34:32 +0000 (10:34 +0900)]
Add explicit padding to conv (#1936)

* Add explicit padding to conv

This commit adds explicit padding to conv.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Revise namespace as Explicit

This commit revises namespace as Explicit for consistency.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago Add explicit padding to average pooling (#1928)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Jul 2018 00:50:08 +0000 (09:50 +0900)]
 Add explicit padding to average pooling (#1928)

* Add explicit padding to average pooling

This commit adds explicit padding to average pooling.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Remove unused and unnecessary blank

This commit removes unused and unnecessary blank

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Revise namespace as Explicit

This commit revises namespace as Explicit for consistency.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[newrt] Implement Concat kernel for CPU (#1924)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 23:46:57 +0000 (08:46 +0900)]
[newrt] Implement Concat kernel for CPU (#1924)

Related : #1860

This commit implements `Concat` kernel for CPU in new runtime.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years agoUpdate tflite file API (#1832)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Jul 2018 09:53:55 +0000 (18:53 +0900)]
Update tflite file API (#1832)

Update tflite file API generated by flatc
Using updated schema in tensorflow 1.9

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoSet pureacl runtime to be the main runtime (#1908)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Jul 2018 09:53:33 +0000 (18:53 +0900)]
Set pureacl runtime to be the main runtime (#1908)

0. do not build legacy(nn) runtime into out/lib
   - install legacy(nn) runtime into out/lib/legacy
1. Trigger puracl runtime build on default
2. install pureacl runtime into out/lib

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years ago[newrt] Implement MaxPool kernel for CPU (#1919)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 08:56:50 +0000 (17:56 +0900)]
[newrt] Implement MaxPool kernel for CPU (#1919)

Related : #1860

This commit implements MaxPool kernel for CPU in new runtime.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years agoAdd QuantizationInfo to TensorInfo in operands that supports QASYMM8_TENSOR (#1893)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 08:35:25 +0000 (17:35 +0900)]
Add QuantizationInfo to TensorInfo in operands that supports QASYMM8_TENSOR (#1893)

This commit adds QuantizationInfo to TensorInfo in operands that supports QASYMM8_TENSOR.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoFix wrong inclusion and scale value in nnapi_unittests. (#1926)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 07:10:52 +0000 (16:10 +0900)]
Fix wrong inclusion and scale value in nnapi_unittests. (#1926)

This commit fixes wrong inclusion and scale value in nnapi_unittests.
List of fixed scale value
  - add
  - conv
  - dconv
  - mul

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoAdd strided_slice_ex_float_10 to skip list (#1910)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Jul 2018 06:56:58 +0000 (15:56 +0900)]
Add strided_slice_ex_float_10 to skip list (#1910)

Add GeneratedTests.strided_slice_ex_float_10 to skip list for linux/tizen

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoIntroduce new UnsupportedOperationCount class (#1876)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 05:30:56 +0000 (14:30 +0900)]
Introduce new UnsupportedOperationCount class (#1876)

- This UnsupportedOperationCount class will be used for distinguishing
whether an operator can be counted on operation counts.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years ago[Refactor] Remove if/else(switch case) of rank for addShapeConstr (#1839)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 05:22:21 +0000 (14:22 +0900)]
[Refactor] Remove if/else(switch case) of rank for addShapeConstr (#1839)

* Remove if/else(switch case) of rank for addShapeConstr

Related : #1828

This commit removes if/else(switch cases) of rank for `addShapeConstr`.
That's because `asTensorShape` which used in `addShapeConstr` parameter includes if/else for processing shapes by each ranks.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Apply #1741 cases

* Update rank check for Cast

6 years agoEnable other type of operands in FC operation (#1895)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 05:19:25 +0000 (14:19 +0900)]
Enable other type of operands in FC operation (#1895)

This commit enables other type of operands in FC operation.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agotools/nnapi_unittest runs as GTest (#1810)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 10 Jul 2018 01:49:32 +0000 (10:49 +0900)]
tools/nnapi_unittest runs as GTest (#1810)

Some unittest have not been modified yet, so these cases are changed to GTest.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoTurn off warning on g++ 6.+ (#1916)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Jul 2018 01:38:47 +0000 (10:38 +0900)]
Turn off warning on g++ 6.+ (#1916)

* Turn off warning on g++ 6.+

Turn off warning by arm cl when we use cross compiler version 6.+

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Check gcc version

Check gcc version to set warning flag

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoAdd testcase for left rank < right rank (#1787)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Jul 2018 01:25:10 +0000 (10:25 +0900)]
Add testcase for left rank < right rank (#1787)

* Add testcase for left rank < right rank

This commit adds a testcase for different rank
-left rank < right rank

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Change to a gtest format

This commit changes main format to a gtest.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoRemove incorrect link_library to neuralnetworks (#1918)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 09:23:20 +0000 (18:23 +0900)]
Remove incorrect link_library to neuralnetworks (#1918)

In neural_networks_test, libneuralnetworks incorrectly is linked.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoEnables other type of operands in CONV2D operation. (#1889)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 9 Jul 2018 07:19:46 +0000 (16:19 +0900)]
Enables other type of operands in CONV2D operation. (#1889)

* Change the type of CONV2D's Bias from QASYMM8 to INT32

This commit changes CONV2D's bias type from QASYMM8 to INT32.

According to the nnapi description, for input tensor of {@link ANEURALNETWORKS_TENSOR_QUANT8_ASYMM} type, the bias should be of {@link ANEURALNETWORKS_TENSOR_INT32}.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
* Add QuantizationInfo to TensorInfo in CONV2D operation.

This commit adds QuantizationInfo to TensorInfo in CONV2D operation.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[New runtime] Skeleton for addOperationEx (#1913)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 06:45:18 +0000 (15:45 +0900)]
[New runtime] Skeleton for addOperationEx (#1913)

Prepare addOperationEx
Fix link error

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoUpdate ACL (Cast float to qasymm8) (#1881)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 9 Jul 2018 06:39:04 +0000 (15:39 +0900)]
Update ACL (Cast float to qasymm8) (#1881)

This commit updates ACL (Cast float to qasymm8).

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoDisable pureacl/new runtine build which depends on arm compute library (#1909)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 05:45:11 +0000 (14:45 +0900)]
Disable pureacl/new runtine build which depends on arm compute library (#1909)

On x64, disable pureacl/new runtine build which depends on arm compute library

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoChange the scale value of unit test related to the convolution. (#1887)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 9 Jul 2018 05:43:23 +0000 (14:43 +0900)]
Change the scale value of unit test related to the convolution. (#1887)

This commit change the scale value of unit test related to the convolution.
- Change the scale value of output operand to 1.0f.
- Change the scale value of input and kernel operand to 0.02f.
- Change the scale value of bias to the value of multiplying input scale and kernel scale.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[newrt] Remove CPU prefix of ConvolutionLayer class (#1891)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 9 Jul 2018 05:23:42 +0000 (14:23 +0900)]
[newrt] Remove CPU prefix of ConvolutionLayer class (#1891)

This commit removes CPU prefix of ConvolutionLayer class

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago--env-file option for docker cross arm build (#1879)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 04:54:00 +0000 (13:54 +0900)]
--env-file option for docker cross arm build (#1879)

* --env-file option for docker cross arm build

1. support --env-file option via $ENV_FILE
2. add env_file for pureacl, new_runtime build

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* Rename env files

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* remove empty line

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoModify wrong values in testcases (#1865)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Jul 2018 02:08:29 +0000 (11:08 +0900)]
Modify wrong values in testcases (#1865)

This commit modifies wrong values in testcases.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[newrt] Fix build failure in > gcc 5.6 (#1898)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 9 Jul 2018 01:45:15 +0000 (10:45 +0900)]
[newrt] Fix build failure in > gcc 5.6 (#1898)

This commit fixes build failure in > gcc 5.6.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years agoRemove test build from newruntime build (#1903)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 01:44:59 +0000 (10:44 +0900)]
Remove test build from newruntime build (#1903)

There is link error when building runtime_run_android_nn_test with new runtime.
Before fixing all buld error by newruntime, disable runtime_run_android_nn_test.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoUpdate NeuralNetworks.h from android-p-prevew-4 (#1905)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Jul 2018 01:37:09 +0000 (10:37 +0900)]
Update NeuralNetworks.h from android-p-prevew-4 (#1905)

Related Issue: #1904

Android-p-preview-3 announced that it will have fixed NNAPI v1.1.
However I would like to update NeuralNetworks.h from android-p-preview-4's.

Because

- it updated comments (style, description, ...).
- it fixed a mistake included in android-p-preview-3.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agolink runtime_run_android_nn_test to nnapi_puer_arm_compute (#1901)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Jul 2018 01:35:38 +0000 (10:35 +0900)]
link runtime_run_android_nn_test to nnapi_puer_arm_compute (#1901)

this commit adds link runtime_run_android_nn_test to nnapi_puer_arm_compute,
which is missing now.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoFix link error when linking runtime_run_android_nn_test with pureacl (#1900)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 6 Jul 2018 08:14:00 +0000 (17:14 +0900)]
Fix link error when linking runtime_run_android_nn_test with pureacl (#1900)

Fix : #1899
1. add dummy APIs to fix link error

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoEnable the kTfLiteUInt8 of the input operands in RandomTestRunner to be random value...
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 6 Jul 2018 06:45:02 +0000 (15:45 +0900)]
Enable the kTfLiteUInt8 of the input operands in RandomTestRunner to be random value. (#1882)

This commit enables the kTfLiteUInt8 of the input operands in RandomTestRunner to be random value.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[newrt] Implement AvgPool kernel for CPU (#1880)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 6 Jul 2018 02:31:51 +0000 (11:31 +0900)]
[newrt] Implement AvgPool kernel for CPU (#1880)

* [newrt] Implement AvgPool kernel for CPU

This commit implements AvgPool kernel for CPU in new runtime.

Signed-off-by:sjsujinkim sjsujin.kim@samsung.com

* Relocate to_string to support/nnapi

6 years agoIntroduce EnvVar class (#1884)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 6 Jul 2018 02:09:14 +0000 (11:09 +0900)]
Introduce EnvVar class (#1884)

* Introduce EnvVar class

EnvVar reads environment variables from system and can return the
value as string, bool or int.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Move EnvVar.h to nnfw/include/util

Move the file and add namespace

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoDefine android nn test skiplist for pure arm compute runtime (#1854)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 6 Jul 2018 01:58:39 +0000 (10:58 +0900)]
Define android nn test skiplist for pure arm compute runtime (#1854)

* Define android nn test skiplist for pure arm compute runtime

1. backup the existing skiplist for legacy runtime
2. define new skiplist for pure arm compute runtime

related issue: #1851

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* rename backup file

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* add new line

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoAdd unit test for operations that support ANEURALNETWORKS_TENSOR_QUANT8_ASYMM (#1857)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 07:30:01 +0000 (16:30 +0900)]
Add unit test for operations that support ANEURALNETWORKS_TENSOR_QUANT8_ASYMM (#1857)

This commit adds unit tests for operations that support ANEURALNETWORKS_TENSOR_QUANT8_ASYMM.

Operation list of added unit test
  - ANEURALNETWORKS_TENSOR_ADD
  - ANEURALNETWORKS_AVERAGE_POOL_2D
  - ANEURALNETWORKS_CAST_EX (QUANT8_ASYMM to FLOAT32)
  - ANEURALNETWORKS_CONCATENATION
  - ANEURALNETWORKS_CONV_2D
  - ANEURALNETWORKS_DEPTHWISE_CONV_2D
  - ANEURALNETWORKS_FULLY_CONNECTED
  - ANEURALNETWORKS_MAX_POOL_2D
  - ANEURALNETWORKS_MUL
  - ANEURALNETWORKS_RELU
  - ANEURALNETWORKS_RELU6
  - ANEURALNETWORKS_RESHAPE
  - ANEURALNETWORKS_SOFTMAX

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoAdd buildscript for new runtime (#1875)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 5 Jul 2018 07:12:08 +0000 (16:12 +0900)]
Add buildscript for new runtime (#1875)

This script is used on CI to build new runtime

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoApply OperatorParser to model_parser.py (#1871)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 06:04:37 +0000 (15:04 +0900)]
Apply OperatorParser to model_parser.py (#1871)

Apply OperatorParser to model_parser.py. Now model_parser.py
uses OperatorParser.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years ago[newrt] Use ACL backend for all operations (#1874)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 05:49:21 +0000 (14:49 +0900)]
[newrt] Use ACL backend for all operations (#1874)

Currently we cannot run with heterogeneous backends, set all backends as
ACL. Later, we are going to control it with environment variables.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoApply format checker to new runtime (#1872)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 05:28:38 +0000 (14:28 +0900)]
Apply format checker to new runtime (#1872)

This commit applies format checker to new runtime

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[CMake] Extract building runtime test as a param (#1870)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 05:14:11 +0000 (14:14 +0900)]
[CMake] Extract building runtime test as a param (#1870)

Before `add_subdirectory(tests)` was called twice if we run both two
runtimes which causes cmake error.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoCreate frameworktest list file for new runtime (#1869)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 5 Jul 2018 05:04:09 +0000 (14:04 +0900)]
Create frameworktest list file for new runtime (#1869)

convolution1 and convolution2 model added

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoIntroduce OperatorParser for model_parser (#1858)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 04:06:20 +0000 (13:06 +0900)]
Introduce OperatorParser for model_parser (#1858)

* Introduce OperatorParser for model_parser

- To remove `TFLiteModelFileParser`'s dependencies for tflite.operator,
    tflite.tensor, introduce new class `OperatorParser`.
- `OperatorParser` class handles `Operator` class and `Tensor` class.
- Not yet changed model_parser.py. To test this, it needs another commit
for new operator counter and changes on model_parser.py

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Rename __ to _wrapping

6 years agoIntroduce OperationCount and PerfPredictor (#1859)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 04:06:03 +0000 (13:06 +0900)]
Introduce OperationCount and PerfPredictor (#1859)

* Introduce OperationCount and PerfPredictor

- For #1836
- OperationCount class has count units and these will be calculated for
total cycle by PerfPredictor
- PerfPredictor class will calculate total cycles with OperationCount.
By setting each cycle's value, performance prediction will be
calculated.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Add more description for how to calc in detail

6 years agoAdd nnapi_unittest for ReLU1 (#1864)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 03:46:23 +0000 (12:46 +0900)]
Add nnapi_unittest for ReLU1 (#1864)

relu1_1 is for 2 ranks input model.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoDisable logistic/l2pool2d on framework test (#1856)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 5 Jul 2018 03:20:42 +0000 (12:20 +0900)]
Disable logistic/l2pool2d on framework test (#1856)

On pure arm compute runtime, logistic/l2pool2d fail on framework test.
This commit makes CI to run seamless when pureacl runtime runs as default runtime.

Related issue : #1851

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoDefinitions for supporting Tanh in pureacl runtime (#1846)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 03:17:39 +0000 (12:17 +0900)]
Definitions for supporting Tanh in pureacl runtime (#1846)

Define namespace, node, param for Tanh

Signed-off-by: seok <sk.namkoong@samsung.com>
6 years agoSupport other types for TfLiteInterpMatchApp (#1842)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 03:14:13 +0000 (12:14 +0900)]
Support other types for TfLiteInterpMatchApp (#1842)

This commit supports other types for TfLiteInterpMatchApp.
- Change TfLiteInterpMatchApp::compareSingleTensorView() to the template method.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoAdd Kernel ACL test case for ReLU6 (#1766)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 5 Jul 2018 01:34:07 +0000 (10:34 +0900)]
Add Kernel ACL test case for ReLU6 (#1766)

Add tests for both NEON and CL kernel were added
Support tests for 2d-tensors and 4d-tensors
To compare a performance btw BOUNDED_RELU and LU_BOUNDED_RELU,
add two kinds of ReLU6 implementation
  - BOUNDED_RELU : Upper Bounded Rectifier (f(x) = min(a, max(0,x))
  - LU_BOUNDED_RELU : Lower and Upper Bounded Rectifier (f(x) = min(a, max(b,x))

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years ago[new_runtime] Support CPU backend (#1847)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 09:32:47 +0000 (18:32 +0900)]
[new_runtime] Support CPU backend (#1847)

Conv2D Operation is only supported yet.
Implement #1607 into the current runtime structure.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoSupport other types for RandomGenerator (#1831)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 07:30:34 +0000 (16:30 +0900)]
Support other types for RandomGenerator (#1831)

This commit supports other types(int32_t, uint8_t) for RandomGenerator.
- Add quantization info to RandomGenerator.
- Change class RandomGenerator to not be the template class.
- Change operator method to template method(generate<>()).

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoAdd ops_counter for operator types in Inceptionv3 (#1824)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 07:22:59 +0000 (16:22 +0900)]
Add ops_counter for operator types in Inceptionv3 (#1824)

Add ops_counter for operator types in Inceptionv3.
- Pooling, Softmax, FullyConnected.
- LogSoftmax: this is added but not used.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years agoAdd inline keyword to `iterate()` (#1848)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 07:11:12 +0000 (16:11 +0900)]
Add inline keyword to `iterate()` (#1848)

To avoid multiple definition error

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[new_runtime] Initial structure change (#1843)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 05:31:01 +0000 (14:31 +0900)]
[new_runtime] Initial structure change (#1843)

Change runtime structure to support multiple backends

- Introduce `Backend` which contains IInitializerGenerator and
  IStageGenerator
- Introduce `BackendResolver` so we can choose which backend for each
  operation
- Introduce `IObject` so it can access any kind of tensors from different
  backends
- Move arm_compute-specific implementation to arm_compute directory
- Keep Concat layer for later elimination since it is hard to know at
  first if we have heterogeneous backends

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoAdd a float variable of nearest to 1 (#1837)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 03:59:49 +0000 (12:59 +0900)]
Add a float variable of nearest to 1 (#1837)

This commit adds a float variable of nearest to 1.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoChange install dir of new runtime (#1844)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 4 Jul 2018 02:47:31 +0000 (11:47 +0900)]
Change install dir of new runtime (#1844)

1. install new runtime into lib/new_runtime

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoRecover the sequence of LD_LIBRARY_PATH (#1838)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 4 Jul 2018 01:36:52 +0000 (10:36 +0900)]
Recover the sequence of LD_LIBRARY_PATH (#1838)

The sequence of LD_LIBRARY_PATH should be remained as it was.

Recover the sequence of LD_LIBRARY_PATH

Signed-off-by: seok <sk.namkoong@samsung.com>
6 years agotflite_benchmark: use get_env_bool instead of getenv (#1812)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 3 Jul 2018 23:54:33 +0000 (02:54 +0300)]
tflite_benchmark: use get_env_bool instead of getenv (#1812)

* tflite_benchmark: use get_env_bool instead of getenv

Replace std::getenv by nnfw::util::get_env_bool for USE_NNAPI

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* Update tflite_benchmark.cc

6 years agotflite_benchmark: get env refactor (#1801)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 3 Jul 2018 23:54:18 +0000 (02:54 +0300)]
tflite_benchmark: get env refactor (#1801)

* tflite_benchmark: get env refactor

using nnfw::util::get_env_int instead of custom get_env call

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* add const spec

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* fix compare signed/unsigned issue

6 years agotflite_benchmark: use get_env_int for THREAD (#1840)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 3 Jul 2018 10:40:08 +0000 (13:40 +0300)]
tflite_benchmark: use get_env_int for THREAD (#1840)

Replace std::getenv by nnfw::util::get_env_int for THREAD

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
6 years agoConsider different rank between left and right operands for broadcasting (#1786)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 3 Jul 2018 10:34:36 +0000 (19:34 +0900)]
Consider different rank between left and right operands for broadcasting (#1786)

This commit resolve different rank between left and right operands for broadcasting

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agofix get_env_bool for "true"/"false" values (#1811)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 3 Jul 2018 09:46:26 +0000 (12:46 +0300)]
fix get_env_bool for "true"/"false" values (#1811)

* fix get_env_bool for "true"/"false" values

exception in case non int string passed

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* remove unused code

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
6 years agoModify Framework test script (#1829)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 3 Jul 2018 05:28:50 +0000 (14:28 +0900)]
Modify Framework test script (#1829)

Setting of NNFW_LD_LIBRARY_PATH is done many times
Even though LD_LIBRARY_PATH is given, script search NNFW_LD_LIBRARY_PATH first.

Put the codes of setting NNFW_LD_LIBRARY_PATH outside of iteration block
Change the sequence of LD_LIBRARY_PATH and NNFW_LD_LIBRARY_PATH

6 years agoSeparate make_default_quantization function. (#1823)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 3 Jul 2018 02:53:18 +0000 (11:53 +0900)]
Separate make_default_quantization function. (#1823)

This commit seperate make_default_quantization to avoid error of redefinition.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoMove ops_counters code to op_counter.py (#1813)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 3 Jul 2018 00:59:07 +0000 (09:59 +0900)]
Move ops_counters code to op_counter.py (#1813)

* Move ops_counters code to op_counter.py

Move ops_counters code to op_counter.py

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Add code to prevent no key sitatuion

* Add operator types for inceptionv3

6 years agoUpdate NeuralNetworks.h: NN API 1.1 (#1815)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 2 Jul 2018 07:35:42 +0000 (16:35 +0900)]
Update NeuralNetworks.h: NN API 1.1 (#1815)

* Update NeuralNetworks.h: NN API 1.1

Update NeuralNetworks.h to released NN API 1.1
https://android.googlesource.com/platform/frameworks/ml/+/android-p-preview-3/nn/runtime/include/NeuralNetworks.h

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Remove andorid api version check

Remove andorid api version check

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Remove base version check

Remove base android version check

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoDo not install nnapi_unittest testcases (#1817)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 2 Jul 2018 07:11:44 +0000 (16:11 +0900)]
Do not install nnapi_unittest testcases (#1817)

These testcases are not an official test tools
So do not install nnapi_unittest testcases to unitteset directory.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoUpdates for cross build using docker (#1807)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 2 Jul 2018 02:26:44 +0000 (11:26 +0900)]
Updates for cross build using docker (#1807)

Modify some sentences
Add method about cross build using docker with ACL build
Remove method for cross build using docker without ACL build
(We don't use method without ACL build anymore)

6 years agoenv: add default value parameter to get_env (#1800)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Mon, 2 Jul 2018 02:02:52 +0000 (05:02 +0300)]
env: add default value parameter to get_env (#1800)

* env: add default value parameter to get_env

defaultValue parameter can be used if named parametr is missing

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* move default value to decaration

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
6 years agoRemove dependency with internal header for custom op (#1804)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 2 Jul 2018 01:50:10 +0000 (10:50 +0900)]
Remove dependency with internal header for custom op (#1804)

Remove dependency with internal header to support custom operation

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoAdds counter for operations of convolution operator (#1803)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 29 Jun 2018 08:19:37 +0000 (17:19 +0900)]
Adds counter for operations of convolution operator (#1803)

* Adds counter for operations of convolution operator

Adds counter for operations of convolution operator
- Ref:
https://github.com/apaszke/torch-opCounter/blob/e610cebf12c7ba36f5ee1c6b327129478202b32e/src/profiler.lua#L75

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Apply format.patch

6 years agoAdd nnapi_unittest for ReLU6 (2D-tensor) (#1794)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 29 Jun 2018 04:04:07 +0000 (13:04 +0900)]
Add nnapi_unittest for ReLU6 (2D-tensor) (#1794)

relu6_1 is for 2 ranks input model.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoUpdate acl patch (gather operation) (#1769)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 29 Jun 2018 00:49:40 +0000 (09:49 +0900)]
Update acl patch (gather operation) (#1769)

This commit updates acl patch.
-https://github.sec.samsung.net/RS7-RuntimeNTools/ComputeLibrary/pull/26
-related with #1746

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoAdd an unit test for TANH op (#1774)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 28 Jun 2018 07:47:11 +0000 (16:47 +0900)]
Add an unit test for TANH op (#1774)

This commit adds an unit test for TANH op.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoUse GLOB in nn runtime's CMakeLists.txt (#1798)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 28 Jun 2018 06:30:09 +0000 (15:30 +0900)]
Use GLOB in nn runtime's CMakeLists.txt (#1798)

Use GLOB for listing all sources becasue ...

- Currently we compile all *.cpp sources.
- By using GLOB, we don't need to modify CMakeLists.txt.
- Our other runtimes (pure_arm_compute, new_runtime) already use this feature.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoSupport ReLU1 operation (#1795)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 28 Jun 2018 06:29:27 +0000 (15:29 +0900)]
Support ReLU1 operation (#1795)

Define ReLU1 namespace, node, param in pure acl runtime
Set parameter for ReLU1 in pure acl runtime

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoAdd runtime information in libneuralnetworks.so (#1793)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 28 Jun 2018 03:36:57 +0000 (12:36 +0900)]
Add runtime information in libneuralnetworks.so (#1793)

Related Issue: #1792

You can get the name of runtime from `libneuralnetworks.so` by command:
```
$ strings Product/out/lib/libneuralnetworks.so         | grep "library info"
library information : runtime=nn

$ strings Product/out/lib/pureacl/libneuralnetworks.so | grep "library info"
library information : runtime=pure_arm_compute
```

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoSupport ReLU6 operation (#1770)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 28 Jun 2018 01:46:58 +0000 (10:46 +0900)]
Support ReLU6 operation (#1770)

Define ReLU6 namespace, node, param in pure acl runtime
Set parameter for ReLU6 in pure acl runtime

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agotflite_benchmark: Fix segfault on model loading. (#1782)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 27 Jun 2018 10:52:04 +0000 (13:52 +0300)]
tflite_benchmark: Fix segfault on model loading. (#1782)

* tflite_benchmark: Fix segfault on model loading.

add BuildFromFile return result checking

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* cout to cerr

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
6 years agoAdd tflite_benchmark console params checking (#1781)
Vitaliy Cherepanov/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 27 Jun 2018 10:51:30 +0000 (13:51 +0300)]
Add tflite_benchmark console params checking (#1781)

* Add tflite_benchmark console params checking

add params count checking
add help

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
* add ostream as param

Signed-off-by: Vitaliy Cherepanov <v.cherepanov@samsung.com>
6 years agoAdd testcases for SUB broadcasting (#1789)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 27 Jun 2018 10:44:31 +0000 (19:44 +0900)]
Add testcases for SUB broadcasting (#1789)

This commit adds testcases for SUB broadcasting
-left > right (rank): test5
-left < right (rank): test6

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoPrint operator types and its count for model (#1790)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 27 Jun 2018 09:25:20 +0000 (18:25 +0900)]
Print operator types and its count for model (#1790)

Print operator types and its count for model

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years agoChange model_parser.py to print tensors of each input/output (#1785)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 27 Jun 2018 02:38:29 +0000 (11:38 +0900)]
Change model_parser.py to print tensors of each input/output (#1785)

* Change model_parser.py to print tensors of each input/output

Change model_parser.py to print tensors of each input/output. This
commit makes users easily see input/output tensors.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* fix format to pass format-checker

6 years agotools/nnapi_unittest runs as GTest (#1757)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 27 Jun 2018 01:02:03 +0000 (10:02 +0900)]
tools/nnapi_unittest runs as GTest (#1757)

Changes each test to GTest.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoFix daily_arm_coverage build break (#1779)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 27 Jun 2018 00:47:20 +0000 (09:47 +0900)]
Fix daily_arm_coverage build break (#1779)

Fix daily_arm_coverage build break : issue #1776

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoAdd nnapi_unittest for ReLU (#1761)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 26 Jun 2018 23:31:19 +0000 (08:31 +0900)]
Add nnapi_unittest for ReLU (#1761)

relu_1 is for 2 ranks input model.
relu_2 is for 3 ranks input model.
relu_3 is for 4 ranks input model.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoExtend broadcasting for ADD operation (#1765)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 26 Jun 2018 10:00:42 +0000 (19:00 +0900)]
Extend broadcasting for ADD operation (#1765)

* Extend broadcating for ADD operation

This commit extends broadcating for ADD operation.
-Only if input are initialized, it supports various broadcasting.
-For example, in case left = {N=1,H=3,W=2,C=4}
    these cases(right={4} or right={244} or right=3,2,4}) are supported.
-Previously, only right={1} is supported.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Remove unused blank and rename variables

This commit removes unused blank and rename variables.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Remove addShapeForBroadcast and revise addShapeConstr

This commit removes addShapeForBroadcast and revises addShapeConstr.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Change it->second.rank() to 4 (constant)

This commit changes it->second.rank() to 4
-because it always 4.
-Broadcasting tensor shape is needed only for 4D input.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Revise an incorrect comment

This commit modifies an incorrect comment.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Update and modify wrong format

This commit updates wrong format.

Signed-off-by: SungJin Choi lotieye.choi@samsung.com
Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Delete duplicate addShapeConstr statement

This commit deletes duplicate addShapeConstr statement.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago Introduce broadcasting testcase for ADD operation (#1768)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 26 Jun 2018 05:36:21 +0000 (14:36 +0900)]
 Introduce broadcasting testcase for ADD operation (#1768)

* Introduce broadcasting cast for ADD operation

This commit itroduces more broadcasting cast for ADD operation.
-to test #1765

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* add a line at the end

This commit adds a line at the end.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoAdd Kernel ACL test case for ReLU (#1763)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 26 Jun 2018 03:53:22 +0000 (12:53 +0900)]
Add Kernel ACL test case for ReLU (#1763)

Add tests for both NEON and CL kernel were added
Support tests for 2d-tensors and 4d-tensors

KernelACL_TC.reluFloat32_2d
KernelACL_TC.reluFloat32_4d
KernelACL_TC.neon_reluFloat32_2d
KernelACL_TC.neon_reluFloat32_4d

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years ago Set parameter for ReLU in pure acl runtime (#1759)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 26 Jun 2018 00:20:39 +0000 (09:20 +0900)]
 Set parameter for ReLU in pure acl runtime  (#1759)

* Set parameter for ReLU in pure acl runtime

Set parameter for ReLU in pure acl runtime compilation step

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Remove irrelevant code

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoImplement DEQUANTIZE operation of NNAPI in pureacl. (#1740)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 25 Jun 2018 05:31:59 +0000 (14:31 +0900)]
Implement DEQUANTIZE operation of NNAPI in pureacl. (#1740)

This commit implements DEQUANTIZE operation in pureacl.
- Enable DEQUANTIZE operator on nnapi_delegate of tflite.
- Implement DEQUANTIZE operation in pureacl

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoSupport general brodcasting on ADD (#1733)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 25 Jun 2018 04:27:38 +0000 (13:27 +0900)]
Support general brodcasting on ADD (#1733)

* Support general brodcasting on ADD

This commit supports general brodcasting on ADD.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Modify comparison between shape

This commit modifies comparison between shape.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoDefine ReLU namespace in pure acl runtime (#1737)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 25 Jun 2018 02:11:34 +0000 (11:11 +0900)]
Define ReLU namespace in pure acl runtime (#1737)

Define ReLU namespace, node, param in pure acl runtime

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoDIV TFLITE file Generator for testing (#1742)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 07:16:45 +0000 (16:16 +0900)]
DIV TFLITE file Generator for testing (#1742)

This file generates the following tflite files for DIV.
These are used for #1741.

```
$ ls -1 models/div/*.tflite

models/div/div_1d_1d.tflite
models/div/div_1d_scalarConst.tflite
models/div/div_2d_1d.tflite
models/div/div_2d_2d.tflite
models/div/div_2d_scalarConst.tflite
models/div/div_3d_1d.tflite
models/div/div_3d_2d.tflite
models/div/div_3d_3d.tflite
models/div/div_3d_scalarConst.tflite
models/div/div_4d_1d.tflite
models/div/div_4d_2d.tflite
models/div/div_4d_3d.tflite
models/div/div_4d_4d.tflite
models/div/div_4d_scalarConst.tflite
models/div/div_scalarConst_scalarConst.tflite
```

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoSupport QASYMM8 type to initializer of CONV2D op and FC op (#1710)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 22 Jun 2018 06:21:07 +0000 (15:21 +0900)]
Support QASYMM8 type to initializer of CONV2D op and FC op (#1710)

This commit supports QASYMM8 type to initializer of CONV2D op and FC op.
- Add initKernelTensor() function.
- Support QASYMM8 type to initializer of CONV2D op and FC op.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[tool] PB info printer (#1749)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 06:14:56 +0000 (15:14 +0900)]
[tool] PB info printer (#1749)

This tool prints information of operations inside a PB file.
Currently this information will  be used to calculate overhead of Conv in ADAS model.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoAdd a unit test for dequantize op. (#1735)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 22 Jun 2018 06:13:49 +0000 (15:13 +0900)]
Add a unit test for dequantize op. (#1735)

This commit adds a unit test for dequantize op.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[Emergency] Fix build break (#1756)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 22 Jun 2018 04:12:08 +0000 (13:12 +0900)]
[Emergency] Fix build break (#1756)

Fix build break in compilation.cc

- build.append()
- scala -> scalar

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoAdd testcases for ADD operation (#1734)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 03:48:38 +0000 (12:48 +0900)]
Add testcases for ADD operation (#1734)

This commit adds test cases for ADD operation to check various broadcasting cases.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoScale for Softmax (#1728)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 03:40:26 +0000 (12:40 +0900)]
Scale for Softmax (#1728)

Parent issue: #1675
This commit enables scale (beta) for `Softmax` operation.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>