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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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
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>
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>
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>
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
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>
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
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>
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>
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)
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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 03:40:00 +0000 (12:40 +0900)]
Introduce ReLU1 to pure acl runtime (#1715)
This commit adds ReLU1 to pure acl runtime.
-min(max(x, -1), 1);
Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 22 Jun 2018 03:38:56 +0000 (12:38 +0900)]
Add activiation to avgpool and maxpool (#1718)
This commit adds activiation to avgpool and maxpool in pure acl runtime.
Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 22 Jun 2018 02:40:27 +0000 (11:40 +0900)]
Move functions related to make TensorInfo to Cast.h (#1705)
* Move functions related to make TensorInfo to Cast.h
This commit moves functions related to make TEnsorInfo from compilation.cc to Cast.h
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
* Add quantization information to an asTensorInfo() for operand::Shape
This commit adds quantization information to an asTensorInfo() for operand::Shape.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>