platform/core/ml/nnfw.git
6 years agoIntroduce View, Reader and IndexIterator of Matrix (#2309)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 16 Aug 2018 07:06:15 +0000 (16:06 +0900)]
Introduce View, Reader and IndexIterator of Matrix (#2309)

Introduce View, Reader and IndexIterator of Matrix. These classes would
be used for initializing constant matrix.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years agoAdd python script of run_frameworktest (#2085)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 16 Aug 2018 04:41:39 +0000 (13:41 +0900)]
Add python script of run_frameworktest (#2085)

This commit will add python script of run_frameworktest

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
6 years agoUpdate ACL (#2301)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 16 Aug 2018 02:26:58 +0000 (11:26 +0900)]
Update ACL (#2301)

- Enable QASYMM8 in the CLArithmeticAdditionKernel.
- Extend the range in which StridedSlice is supported for CPU.
- Bugfix of subtensor validation.
- Unify the order of input data in StridedSliceCPU.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years ago[neurun] Introduce Graph Phase (#2298)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 10:14:11 +0000 (19:14 +0900)]
[neurun] Introduce Graph Phase (#2298)

`graph::Graph::Phase` is an enum class that represents which phase the
Graph is at. Since we are going to use `Graph` class for multiple
phases we need this with verification infra.

- Introduce an enum class `graph::Graph::Phase`
- Phase transition when `ANeuralNetworksModel_finish()` is called
- Call verifier from the appropriate place

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Introduce IVerifier interface (#2296)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 09:00:21 +0000 (18:00 +0900)]
[neurun] Introduce IVerifier interface (#2296)

Introduce IVerifier interface to support various verifications, each
verification can be reused from different graph phases.

This commit also contains a skeleton of `DAGChecker`.

Part of #2247

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Extract TensorMarker to codegen (#2294)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 07:54:53 +0000 (16:54 +0900)]
[neurun] Extract TensorMarker to codegen (#2294)

Extract TensorMarker into a separate file and move it to codegen
directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Move operand IObject to backend (#2293)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 07:40:26 +0000 (16:40 +0900)]
[neurun] Move operand IObject to backend (#2293)

Move `internal::Object` to `neurun::backend::operand::Object`.

Part of #2286

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Move cpu operand Object to backend (#2289)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 04:52:09 +0000 (13:52 +0900)]
[neurun] Move cpu operand Object to backend (#2289)

Move `internal::cpu::operand::Object` to
`neurun::backend::cpu::operand::Object`.

Part of #2286

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Fix feature to ACL TensorShape conversion (#2290)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 14 Aug 2018 04:39:28 +0000 (13:39 +0900)]
[neurun] Fix feature to ACL TensorShape conversion (#2290)

Fix feature to ACL TensorShape conversion to set batch value

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Extract Planner to codegen (#2291)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 03:47:16 +0000 (12:47 +0900)]
[neurun] Extract Planner to codegen (#2291)

Extract Planner into a separate file and move it to codegen directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoFix for Max_pool_float_* Gtest Failure (#2283)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 14 Aug 2018 02:58:10 +0000 (08:28 +0530)]
Fix for Max_pool_float_* Gtest Failure (#2283)

This is fix for max_pool_float_*.Batches are now set correctly in arm_compute::tensorshape

Signed-off-by: Shubham <shub98.gupta@samsung.com>
6 years ago[neurun] Extract asXX helpers into seperate file (#2278)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 02:08:25 +0000 (11:08 +0900)]
[neurun] Extract asXX helpers into seperate file (#2278)

* [neurun] Extract asXX casting helpers into Cast.h

This commit extracts asXX casting helpers into Cast.h likes PACL(https://github.sec.samsung.net/STAR/nnfw/blob/master/runtimes/pure_arm_compute/src/internal/arm_compute/Cast.h)

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Change file name/add namespace/seperate source file

* Add empty line

6 years agoRename `kernel/cpufallback` to `kernel/cpu` (#2282)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 02:04:07 +0000 (11:04 +0900)]
Rename `kernel/cpufallback` to `kernel/cpu` (#2282)

Rename the directory to remove inconsistency with namespace name in
the code and backend module(cpu backend is in `backend/cpu`).

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Remove unused internal::Model (#2284)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 02:03:32 +0000 (11:03 +0900)]
[neurun] Remove unused internal::Model (#2284)

Now that we replace Model with Graph IR, Model is unused anymore.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Move acl_cl operand Object to backend (#2285)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 14 Aug 2018 01:42:49 +0000 (10:42 +0900)]
[neurun] Move acl_cl operand Object to backend (#2285)

Move `internal::arm_compute::operand::Object` to
`neurun::backend::acl_cl::operand::Object`.

Part of #2286

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoFix unittest script to report error (#2280)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 14 Aug 2018 00:27:51 +0000 (09:27 +0900)]
Fix unittest script to report error (#2280)

Fix unittest script to report error
- Save test result immediately to temporary variable
- We should save result before compare statement

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Remove useless assertion (#2279)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 13 Aug 2018 11:32:20 +0000 (20:32 +0900)]
[neurun] Remove useless assertion (#2279)

Remove useless assertion
It makes validation check fail on debug build

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Introduce IO::Index (#2272)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 13 Aug 2018 10:23:30 +0000 (19:23 +0900)]
[neurun] Introduce IO::Index (#2272)

* [neurun] Introduce IOIndex/IOIndexSet

Close : #2257

This commit introduce IOIndex/IOIndexSet.
`IOIndex` is the indices for ANeuralNetworksModel_identifyInputsAndOutputs() API.
`IOIndexSet` is accessible only using with `IOIndex`.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Revert some codes

6 years agoTest driver: Test all unittest option (#2274)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 13 Aug 2018 10:23:19 +0000 (19:23 +0900)]
Test driver: Test all unittest option (#2274)

Add new option to run all unittest
- Ignore skiplist
- Get test list using gtest option
- Test each unittest item using filter

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Remove unused TensorSetter define (#2275)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 13 Aug 2018 10:23:05 +0000 (19:23 +0900)]
[neurun] Remove unused TensorSetter define (#2275)

This commit removes unused `TensorSetter` function define.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[neurun] Extract PlanBuilder into a separate file (#2276)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 13 Aug 2018 10:22:31 +0000 (19:22 +0900)]
[neurun] Extract PlanBuilder into a separate file (#2276)

Extract PlanBuilder into a separate file and move it to codegen
directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Extract BackendResolver into a separate file (#2273)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 13 Aug 2018 09:17:07 +0000 (18:17 +0900)]
[neurun] Extract BackendResolver into a separate file (#2273)

Extract BackendResolver into a separate file and move it to codegen
directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoIntroduce reduce mean operation for pure acl runtime (#2256)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 13 Aug 2018 09:11:49 +0000 (18:11 +0900)]
Introduce reduce mean operation for pure acl runtime (#2256)

* Rebase on master instead of merging

This commit does rebase on master instead of merging.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* Fix and modify a typo

This commit fixes a typo
- axises => axis

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[neurun] Extract IPlanBuilder into a separate file (#2261)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 13 Aug 2018 01:58:26 +0000 (10:58 +0900)]
[neurun] Extract IPlanBuilder into a separate file (#2261)

Extract IPlanBuilder into a separate file and move it to codegen
directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Move directory of backend interfaces (#2250)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 10 Aug 2018 07:46:18 +0000 (16:46 +0900)]
[neurun] Move directory of  backend interfaces (#2250)

Move backend interfaces to `backend` from `internal`.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoUpdate proxy settings for arm cross build docs (#2255)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 10 Aug 2018 06:23:19 +0000 (15:23 +0900)]
Update proxy settings for arm cross build docs (#2255)

This will update proxy settings for arm cross build documents

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
6 years agoRemove unused apt_proxy (#2253)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 10 Aug 2018 06:22:33 +0000 (15:22 +0900)]
Remove unused apt_proxy (#2253)

This will remove not used anymore apt_proxy file while building arm rootfs

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
6 years ago[neurun] Apply Graph IR (#2244)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 10 Aug 2018 01:52:57 +0000 (10:52 +0900)]
[neurun] Apply Graph IR (#2244)

This commit replaces `internal::tflite::Model` with
`neurun::graph::Graph`.

Resolve #2155

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoSpecify setproxy option in build_rootfs (#2053)
PRAVEEN DORESWAMY NAIDU/System SW /SRI-Bangalore/Staff Engineer/삼성전자 [Fri, 10 Aug 2018 00:20:48 +0000 (05:50 +0530)]
Specify setproxy option in build_rootfs (#2053)

Specify setproxy option with IP address value as argument instead of skipproxy option.
The ipaddress specified with be the actual proxy server.

example: ./build_rootfs.sh setproxy=<IP address of the proxy server>

Signed-off-by: Praveen D N <praveen.dn@samsung.com>
6 years agoDon't formatcheck tools/tflite_benchmark_model with .FORMATDENY (#2242)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 9 Aug 2018 11:01:58 +0000 (20:01 +0900)]
Don't formatcheck tools/tflite_benchmark_model with .FORMATDENY (#2242)

Use .FORMATDENY (#2238) to skip formatcheck tools/tflite_benchmark_model.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years ago[neurun] Extract Plan and related classes from arm_compute namespace (#2239)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 09:59:59 +0000 (18:59 +0900)]
[neurun] Extract Plan and related classes from arm_compute namespace (#2239)

This commit extracts Plan and related classes from arm_compute namespace into ::internal.

6 years agoFix setting in config_aarch64-linux.cmake (#2245)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 09:44:48 +0000 (18:44 +0900)]
Fix setting in config_aarch64-linux.cmake (#2245)

- aarch64-linux-gnu-g*5 -> aarch64-linux-gnu-g*
- proper rootfs path

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years agoModify the quantization info to pass pureacl runtime in broadcast case. (#2098)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 09:05:36 +0000 (18:05 +0900)]
Modify the quantization info to pass pureacl runtime in broadcast case. (#2098)

This commit modifies the quantization info to pass pureacl runtime in broadcast case.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoMake format-checker pushd/popd quiet (#2241)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 07:58:20 +0000 (16:58 +0900)]
Make format-checker pushd/popd quiet (#2241)

Now that we have too many pushd/popd logs this commit makes it not to
print any messages from pushd/popd.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Fix test directory structure (#2240)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 07:54:14 +0000 (16:54 +0900)]
[neurun] Fix test directory structure (#2240)

Move graph test sources to `test/graph` so it will look identical to
`src` directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoIntroduce .FORMATDENY (#2238)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 9 Aug 2018 07:07:58 +0000 (16:07 +0900)]
Introduce .FORMATDENY (#2238)

Put .FORMATDENY in directory that you don't want to check style.

Note: Please don't make ambiguous situation.

e.g.)

```
- A
 +- .FORMATCHECKED
 +- B
    +- FORMATDENY
    +- C
       +- FORMATCHECKED
```

Related issue: #2233

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoUpdate test_driver.py script (#2210)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 06:52:30 +0000 (15:52 +0900)]
Update test_driver.py script (#2210)

Replace `optparse` to `argparse`
Replace `os._exit()` to `sys.exit()`
Remove `TODO : Handle Return Value`
Add exception handling for `artifactpath`
Add `reportdir` parameter and modify related codes
Modify some conditions of checking parameters

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
6 years agoAdd tflite_benchmark_model sources (#2231)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 9 Aug 2018 06:17:20 +0000 (15:17 +0900)]
Add tflite_benchmark_model sources (#2231)

Sources are copied from tensorflow/contrib/lite/tools/benchmark from
v1.9.0-rc2. Then, I modified the followings:

- Namespace is changed
  (tflite -> nnfw)
- Header guard is also changed
- Header include path is adjusted
  (include tensorflow/contrib/lite/profiling/* -> include *)
- Add proper qualification to fix build error (string -> std::string)
- Prepend license for sources under tflite_benchmark_model

In addition, tools/CMakeLists.txt is modified to build tflite_benchmark_model
when `BUILD_TFLITE_BENCHMARK_MODEL` is defined.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoRemove squeeze from skiplist (#2236)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 9 Aug 2018 06:17:01 +0000 (15:17 +0900)]
Remove squeeze from skiplist (#2236)

- Remove squeeze from skiplist

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years ago[neurun] Graph : Introduce Graph (#2235)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 06:16:08 +0000 (15:16 +0900)]
[neurun] Graph : Introduce Graph (#2235)

This commit introduces `neurun::graph::Graph` which represents our new
Graph IR. Note that nodes in the graph is operations.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoSupport QASYMM8 in relu-related operators (#2237)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 05:49:47 +0000 (14:49 +0900)]
Support QASYMM8 in relu-related operators (#2237)

* Support QASYMM8 in relu-related operators

ReLU, ReLU1 and ReLU6 supports quantized operation
It adds QuantizationInfo to TensorInfo in operations that supports QASYMM8_TENSOR

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Enable quant8 testcases of relu-related operator in GeneratedTests

Enable below GeneratedTests:
GeneratedTests.relu_quant8_1
GeneratedTests.relu_quant8_2
GeneratedTests.relu1_quant8_1
GeneratedTests.relu1_quant8_2
GeneratedTests.relu6_quant8_1
GeneratedTests.relu6_quant8_2

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoAdd Logistic operator in pure acl runtime (#2230)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 9 Aug 2018 05:44:01 +0000 (14:44 +0900)]
Add Logistic operator in pure acl runtime (#2230)

* Add Logistic operator in pure acl runtime

Define Logistic namespace, node and param
Set parameter for Logistic in compilation step

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Enable logistic_float_* in GeneratedTests

Enable below two GeneratedTests:
logistic_float_1
logistic_float_2

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
6 years agoEnable squeeze for rank-4 input (#2234)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 9 Aug 2018 04:45:52 +0000 (13:45 +0900)]
Enable squeeze for rank-4 input (#2234)

- Remove input rank restriction from compilation

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agotflite_benchmark_model will not be formatchecked (#2232)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 9 Aug 2018 03:01:11 +0000 (12:01 +0900)]
tflite_benchmark_model will not be formatchecked (#2232)

tflite_benchmark_model will keep original source's style
so that it could be tracked and diff'ed easily.

.FORMATCHEKED in tools (the parent directory of tflite_benchmark_model)
is distributed to all child directories except for
tflite_benchmark_model.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoIntroduce BUILD_TFLITE_BENCHMARK_MODEL option (#2226)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 9 Aug 2018 01:21:39 +0000 (10:21 +0900)]
Introduce BUILD_TFLITE_BENCHMARK_MODEL option (#2226)

Related Issue: #1760, #1696

`tflite_benchmark_model` is a tool for profiling a given model.
The tools is under development in tensorflow upstream.

It was not shipped in v1.9.0 release.
Thus I made `tflite_benchmark_model` work for tflite interpreter
by chery-picking needed patches. (#2212)

This patch is a part of full changes. It only introduces build option.

By default, tflite_benchmark_model will not be built.

To build tflite_benchmark_model tool, use

```
$ OPTIONS='-DBUILD_PURE_ARM_COMPUTE=ON -DBUILD_TFLITE_BENCHMARK_MODEL=ON' \
CROSS_BUILD=1 TARGET_ARCH=armv7l make all install
```

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years ago[neurun] Define operand usage enum and check exception (#2213)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 8 Aug 2018 22:18:15 +0000 (07:18 +0900)]
[neurun] Define operand usage enum and check exception (#2213)

Define operand usage: model input, constant, operation output
Check exception: operand usage conflict

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Graph : Introduce operation Concat (#2223)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:16:50 +0000 (07:16 +0900)]
[neurun] Graph : Introduce operation Concat (#2223)

Introduce `neurun::graph::operation::Concat` which is a node in Graph
IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation Conv2D (#2222)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:16:39 +0000 (07:16 +0900)]
[neurun] Graph : Introduce operation Conv2D (#2222)

Introduce `neurun::graph::operation::Conv2D` which is a node in Graph
IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation MaxPool2D (#2220)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:16:29 +0000 (07:16 +0900)]
[neurun] Graph : Introduce operation MaxPool2D (#2220)

Introduce `neurun::graph::operation::MaxPool2D` which is a node in
Graph IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation Reshape (#2219)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:16:16 +0000 (07:16 +0900)]
[neurun] Graph : Introduce operation Reshape (#2219)

Introduce `neurun::graph::operation::Reshape` which is a node in Graph
IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation FullyConnected (#2221)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:16:05 +0000 (07:16 +0900)]
[neurun] Graph : Introduce operation FullyConnected (#2221)

Introduce `neurun::graph::operation::FullyConnected` which is a node
in Graph IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation Softmax (#2218)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 22:15:49 +0000 (07:15 +0900)]
[neurun] Graph : Introduce operation Softmax (#2218)

Introduce `neurun::graph::operation::Softmax` which is a node in Graph
IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation `Set` (#2215)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 11:07:23 +0000 (20:07 +0900)]
[neurun] Graph : Introduce operation `Set` (#2215)

`neurun::graph::operation::Set` holds operations and also has
ownership of them.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operation AvgPool2D (#2217)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 10:43:36 +0000 (19:43 +0900)]
[neurun] Graph : Introduce operation AvgPool2D (#2217)

Introduce `neurun::graph::operation::AvgPool2D` which is a node in Graph
IR.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agosupport rank 4 feature in pacl runtime (#2183)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 8 Aug 2018 09:37:03 +0000 (18:37 +0900)]
support rank 4 feature in pacl runtime (#2183)

* support rank 4 feature in pacl runtime

This commit enables rank 4 feature

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* rank4 feature for neurun

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years ago[neurun] Remove kernel source from build (#2216)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 09:33:42 +0000 (18:33 +0900)]
[neurun] Remove kernel source from build (#2216)

SOURCES_KERNEL is an empty variable.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Remove unused includes for Initializer in compilation.cc (#2214)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 08:38:57 +0000 (17:38 +0900)]
[neurun] Remove unused includes for Initializer in compilation.cc (#2214)

This commit removes unused includes for Initializer in compilation.cc

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[neurun] Graph : Introduce operand IndexSet (#2209)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 08:36:25 +0000 (17:36 +0900)]
[neurun] Graph : Introduce operand IndexSet (#2209)

`neurun::graph::operand::IndexSet` is a list of `Index`.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoUpdate tensorflow submodule for tflite_benchmark_model (#2212)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 8 Aug 2018 08:28:44 +0000 (17:28 +0900)]
Update tensorflow submodule for tflite_benchmark_model (#2212)

tflite_benchmark_model in v1.9.0 does not build nor work.
This patch fixes build failure with two cherry-picked patches.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years ago[neurun] Graph : Introduce operand Set (#2208)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 08:26:20 +0000 (17:26 +0900)]
[neurun] Graph : Introduce operand Set (#2208)

`neurun::graph::operand::Set` holds operands and also has ownership of
them.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce Operation `Node` (#2206)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 08:24:15 +0000 (17:24 +0900)]
[neurun] Graph : Introduce Operation `Node` (#2206)

Class `neurun::graph::operation::Node` is a wrapper class of
`internal::tflite::op::Node` with input/output tensor info.
Input/Output info will be used for graph traversing.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Set input/output as common tensor,Implement tensor conversion kernels (...
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 07:54:33 +0000 (16:54 +0900)]
[neurun] Set input/output as common tensor,Implement tensor conversion kernels (#2211)

Related : #2142, #2040

- Set input/output as common tensor
- Implement tensor conversion kernels

Tests are matched.

Signed-off-by: sjsujinkim sjsujin.kim@samsung.com
6 years agoRemove newruntime_frameworktest_list.txt (#2207)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 8 Aug 2018 06:53:53 +0000 (15:53 +0900)]
Remove newruntime_frameworktest_list.txt (#2207)

Remove newruntime_frameworktest_list.txt file
It is not used any more

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Change parameters for TensorConversion kernels (#2187)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 05:55:20 +0000 (14:55 +0900)]
[neurun] Change parameters for TensorConversion kernels (#2187)

Related : #2142

This commit changes parameters for TensorConversion kernels.

- Introduce Tensor Shape parameter for iterating tensors by shape info
- Change CPU Data parameter from a pointer to Tensor

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[neurun] Graph : Introduce Operand Index (#2205)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 05:07:52 +0000 (14:07 +0900)]
[neurun] Graph : Introduce Operand Index (#2205)

Introduce `neurun::graph::operand::Index` which is an alias of
`neurun::graph::Index`.

Note that, with the additional template parameter `DummyTag`, aliased
types are treated as different types(strong typedef).

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Graph : Introduce Operation Index (#2204)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 05:07:28 +0000 (14:07 +0900)]
[neurun] Graph : Introduce Operation Index (#2204)

Introduce `neurun::graph::operation::Index` which is an alias of
`neurun::graph::Index`.

Note that, with the additional template parameter `DummyTag`, aliased
types are treated as different types(strong typedef).

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Introduce Index (#2200)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 04:13:48 +0000 (13:13 +0900)]
[neurun] Introduce Index (#2200)

Class `Index` is an integer index wrapper which will be used for
operand, operation and so on.
With template parameter `DummyTag` we can reuse code for many kinds of
indices without adding any code.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Build graph sources (#2201)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 04:13:41 +0000 (13:13 +0900)]
[neurun] Build graph sources (#2201)

Revise CMakeLists.txt to build source files in `graph` directory.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Handle more operand type (#2190)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 8 Aug 2018 02:19:34 +0000 (11:19 +0900)]
[neurun] Handle more operand type (#2190)

- Handle more operand type: ANEURALNETWORKS_TENSOR_INT32, ANEURALNETWORKS_TENSOR_QUANT8_ASYMM
- Handle exception case related with type
- Update unittest skiplist

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Allow to have multi objects in an one index. (#2182)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 02:19:20 +0000 (11:19 +0900)]
[neurun] Allow to have multi objects in an one index. (#2182)

This commit allows to have multi objects in an one index.

It's for objects of heterogeneous backends in an one index.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[neurun] Setup unittest infra (#2191)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 8 Aug 2018 02:19:04 +0000 (11:19 +0900)]
[neurun] Setup unittest infra (#2191)

Setup unittest infra with GTest with a sample test.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[neurun] Use std::nothrow for exception handling (#2184)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 7 Aug 2018 10:43:45 +0000 (19:43 +0900)]
[neurun] Use std::nothrow for exception handling (#2184)

Use std::nothrow for
- Handle exception
- Set returned reference by parameter nullptr

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Mark FromCommon for all of ifm indexes in concat (#2185)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 7 Aug 2018 10:43:35 +0000 (19:43 +0900)]
[neurun] Mark FromCommon for all of ifm indexes in concat (#2185)

This commit marks FromCommon for all of ifm indexes in concat.

It'was a bug.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years agoTest driver: report directory (#2188)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 7 Aug 2018 08:56:27 +0000 (17:56 +0900)]
Test driver: report directory (#2188)

Add parameter to set report directory

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoUpdate unittest skiplist: validation test (#2181)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 7 Aug 2018 08:54:34 +0000 (17:54 +0900)]
Update unittest skiplist: validation test (#2181)

Update validation unittest skiplist
- Remove CreateModel in pacl & neurun
- Remove Model.finish in neurun

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Model finish flag (#2176)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 7 Aug 2018 01:36:45 +0000 (10:36 +0900)]
[neurun] Model finish flag (#2176)

Introduce model finish flag to check duplicated finish call & modification after finished

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoIntroduce docker-shell and docker-run (#2177)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 7 Aug 2018 00:42:53 +0000 (09:42 +0900)]
Introduce docker-shell and docker-run (#2177)

Change script name docker_runs.sh to docker-shell
Introduce docker-run
These scripts' name are same as nncc

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[pacl] NNAPI frontend exception handling by nullptr (#2164)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 6 Aug 2018 05:45:12 +0000 (14:45 +0900)]
[pacl] NNAPI frontend exception handling by nullptr (#2164)

Handle exception by nullptr parameter in NNAPI frontend implementation
Same implementation as neurun

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Fix convolution cpu backend debug build bug (#2150)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 3 Aug 2018 07:00:34 +0000 (16:00 +0900)]
[neurun] Fix convolution cpu backend debug build bug (#2150)

Fix convolution cpu backend debug build bug by check code in tflite kernel

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoUpdate TestValidation.cpp (#2161)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 3 Aug 2018 07:00:21 +0000 (16:00 +0900)]
Update TestValidation.cpp (#2161)

Update TestValication.cpp from android code
Update skiplist

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoFix for Android build (#2160)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 3 Aug 2018 06:05:00 +0000 (15:05 +0900)]
Fix for Android build (#2160)

This will for for Android build to reflect recent changes

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
6 years ago[neurun] Implement StageGenerator for TensorConvert nodes (#2151)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 3 Aug 2018 05:51:43 +0000 (14:51 +0900)]
[neurun] Implement StageGenerator for TensorConvert nodes (#2151)

This commit implements StageGenerator for TensorConvert nodes.

Signed-off-by: sjsujinkim sjsujin.kim@samsung.com
6 years ago[neurun] NNAPI frontend exception handling by nullptr (#2157)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 3 Aug 2018 04:37:45 +0000 (13:37 +0900)]
[neurun] NNAPI frontend exception handling by nullptr (#2157)

Handle exception by nullptr parameter in NNAPI frontend implementation

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Split execution NNAPI implementation into frontend (#2152)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 3 Aug 2018 02:16:01 +0000 (11:16 +0900)]
[neurun] Split execution NNAPI implementation into frontend (#2152)

Split exeution NNAPI implementation into frontend
Move Sink and Source class into internal

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Bug fix: ANeuralNetworksCompilation_finish (#2153)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 10:35:01 +0000 (19:35 +0900)]
[neurun] Bug fix: ANeuralNetworksCompilation_finish (#2153)

Remove duplicated return in ANeuralNetworksCompilation_finish

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] Introduce Common Operands in Plan (#2144)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 2 Aug 2018 07:32:44 +0000 (16:32 +0900)]
[neurun] Introduce Common Operands in Plan (#2144)

Related : #2142

This commit introduces common operands in Plan, since we should set input/output as Common Tensor.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
6 years ago[neurun] Insert TensorConvert nodes (#2127)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 2 Aug 2018 07:29:37 +0000 (16:29 +0900)]
[neurun] Insert TensorConvert nodes (#2127)

* [neurun] Insert TensorConvert nodes

This commit inserts TensorConvert nodes. Those will be inserted to between all of nodes.

Signed-off-by: sjsujinkim sjsujin.kim@samsung.com
* Change to plugin type

* Fix build failure

6 years agoDisable submodule update for obs build (#2149)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 07:01:37 +0000 (16:01 +0900)]
Disable submodule update for obs build (#2149)

Submodule update is disabled in obs build

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoAdd logging of input/output tensors of Mul_1 quicktest (#2140)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 2 Aug 2018 05:42:02 +0000 (14:42 +0900)]
Add logging of input/output tensors of Mul_1 quicktest (#2140)

By this commit, Mul_1 quicktest logs input/output tensor values into specified log file.
The log file is used to create a new Generated File.

Related PR: #2100 (Logger),
            #2105 (#2105 used log file data of this PR to create a new generated test)

Please refer to the comment in #2100 to see how to convert quicktest to generated test.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoUpdate tensorflow submodule (#2143)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 05:05:01 +0000 (14:05 +0900)]
Update tensorflow submodule (#2143)

Update tensorflow submodule to update nnapi_delegate and NeuralNetworksEx.h

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] split compilation NNAPI implementation into frontend (#2145)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 04:55:46 +0000 (13:55 +0900)]
[neurun] split compilation NNAPI implementation into frontend (#2145)

Split compilation NNAPI implementation into frontend
Define ANeuralNetworksCompilation::finish() to divide NNAPI implementation with internal

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoEnvironment variable to set all configurable operators' backend (#2141)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 04:08:24 +0000 (13:08 +0900)]
Environment variable to set all configurable operators' backend (#2141)

Introduce new environment variable to set all configurable operators' backend: OP_BACKEND_ALLOPS

none: use each operator's backend configuration (default)
acl_cl: use acl for all configurable operators' backend
cpu: use cpu for all configurable operators' backend

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[neurun] change logging environment variable name (#2138)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 01:42:49 +0000 (10:42 +0900)]
[neurun] change logging environment variable name (#2138)

Change logging environment variable name: NEURUN_LOG_ENABLE

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoDetect framework test failure (#2137)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 2 Aug 2018 00:36:55 +0000 (09:36 +0900)]
Detect framework test failure (#2137)

Return exit code in run_test.sh to detect framework test error in run_frameworktest.sh
Fix bug (by additional space) in run_frameworktest.sh

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoBug fix: support strided slice (#2136)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 1 Aug 2018 08:11:37 +0000 (17:11 +0900)]
Bug fix: support strided slice (#2136)

Support strided slice operator
STRIDED_SLICE_EX is removed

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoMove memory and model NNAPI implementation to frontend (#2133)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 1 Aug 2018 07:41:25 +0000 (16:41 +0900)]
Move memory and model NNAPI implementation to frontend (#2133)

- Move NNAPI implementation in model.cc into frontend
- Move NNAPI implementation in memory.cc into frontend

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoUpdate NeuralNetworksEx.h (#2134)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 1 Aug 2018 07:39:15 +0000 (16:39 +0900)]
Update NeuralNetworksEx.h (#2134)

- Remove redundant operator
- Remove unittest
- Remove implementation in runtime
- Add SPLIT operator

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoMore `mul` tests for generated test (#2105)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 1 Aug 2018 07:02:18 +0000 (16:02 +0900)]
More `mul` tests for generated test (#2105)

1. converts quick tests for adas to generated test
2. remove successful tests from skip files

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoIntroduce frontend directory (#2130)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 1 Aug 2018 02:59:33 +0000 (11:59 +0900)]
Introduce frontend directory (#2130)

- Make frontend directory
- Move event.cc into frontend
- Modify cmake to build frontend

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoLogging of input and output tensor of quicktest (#2100)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 1 Aug 2018 01:25:33 +0000 (10:25 +0900)]
Logging of input and output tensor of quicktest (#2100)

This is for #2047.
By providing `tensor_logging` param to `RandomTestRunner`,
input and output tensor of quicktest will be saved into specified file.

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