platform/core/ml/nnfw.git
5 years ago[loco] Replace Pad.h with Padding2D.h (#7440)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 16 Sep 2019 03:26:53 +0000 (12:26 +0900)]
[loco] Replace Pad.h with Padding2D.h (#7440)

This commit replaces Pad.h with Padding2D.h in loco.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years ago[moco-tf] Use generated opname (#7435)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 16 Sep 2019 02:27:19 +0000 (11:27 +0900)]
[moco-tf] Use generated opname (#7435)

This commit revises TFFormattedGraph to use generated opname instead of
hardcoded one.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[moco-tf] Introduce TFDialect for Mean (#7320)
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 16 Sep 2019 01:44:48 +0000 (10:44 +0900)]
[moco-tf] Introduce TFDialect for Mean (#7320)

* [moco-tf] Introduce TFDialect for Mean

This commit will introduce TFDialect for `Mean`

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
* apply comments

5 years agoMake to support PReLU op for acl neon (#7330)
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 16 Sep 2019 01:32:37 +0000 (10:32 +0900)]
Make to support PReLU op for acl neon (#7330)

This commit makes to support PReLU op for acl neon.
  - Add PReLU kernel of acl neon
  - Make neurun supports PReLU op for acl neon

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years ago[loco] Rename Pad class to Padding2D (#7395)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 16 Sep 2019 01:30:38 +0000 (10:30 +0900)]
[loco] Rename Pad class to Padding2D (#7395)

This commit is for renaming existing Pad class to Padding2D in loco.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years agoEnable QASYMM8 of PReLU op for acl cl (#7292)
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 16 Sep 2019 01:24:31 +0000 (10:24 +0900)]
Enable QASYMM8 of PReLU op for acl cl (#7292)

This commit enables QASYMM8 of PReLU op for acl cl.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years ago[exo-tflite] Adding MaxPool2DConverter into convert.cpp (#7400)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 16 Sep 2019 00:35:16 +0000 (09:35 +0900)]
[exo-tflite] Adding MaxPool2DConverter into convert.cpp (#7400)

Now MaxPool2DConverter will work with knob on.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[nnc] Remove frontend passes (#7423)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Fri, 13 Sep 2019 20:02:36 +0000 (23:02 +0300)]
[nnc] Remove frontend passes (#7423)

Remove frontend passes. Call importers directly instead.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[nnc] Remove frontend passes in tests (#7424)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Fri, 13 Sep 2019 18:47:44 +0000 (21:47 +0300)]
[nnc] Remove frontend passes in tests (#7424)

Remove frontend passes. Call importers directly instead.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[nnc] Remove frontend passes in utils (#7425)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Fri, 13 Sep 2019 18:44:29 +0000 (21:44 +0300)]
[nnc] Remove frontend passes in utils (#7425)

Remove frontend passes. Call importers directly instead.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir] Remove DTYPE (#7422)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Fri, 13 Sep 2019 16:51:39 +0000 (19:51 +0300)]
[mir] Remove DTYPE (#7422)

Remove no longer used type alias `DTYPE`.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_onnx] Replace DTYPE with DataType. (#7421)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Fri, 13 Sep 2019 14:44:19 +0000 (17:44 +0300)]
[mir_onnx] Replace DTYPE with DataType. (#7421)

Replace deprecated `DTYPE` with `DataType`.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_onnx] Eliminate TransposeOp in Gemm when weights is ConstantOp (#7416)
Павел Ильютченко/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 17:04:41 +0000 (20:04 +0300)]
[mir_onnx] Eliminate TransposeOp in Gemm when weights is ConstantOp (#7416)

* Fixed Gemm importer for elimination extra Transposes

Signed-off-by: Pavel Iliutchenko <p.iliutchenk@samsung.com>
5 years ago[mir] Remove default value of `data_format` parameter (#7420)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 17:00:29 +0000 (20:00 +0300)]
[mir] Remove default value of `data_format` parameter (#7420)

* Remove default value of `data_format` parameter of conv-like op constructors.
* Rename `CommonProps.h` -> `PaddingType.h`.
* Remove deprecated enum members of `PaddingType`.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_tflite] Set `data_format` parameter explicitly (#7415)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 15:19:24 +0000 (18:19 +0300)]
[mir_tflite] Set `data_format` parameter explicitly (#7415)

* Pass `data_format` to constructors of conv-like ops explicitly.
* Replace deprecated `PaddingType::Same` with equivalent `PaddingType::SameUpper`.
* Remove unused includes.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_caffe] Remove unused include (#7417)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 15:15:32 +0000 (18:15 +0300)]
[mir_caffe] Remove unused include (#7417)

Remove one unused include directive

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_caffe2] Remove unused include (#7418)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 15:15:13 +0000 (18:15 +0300)]
[mir_caffe2] Remove unused include (#7418)

Remove one unused include directive

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir2loco] Set data format explicitly (#7419)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 15:14:35 +0000 (18:14 +0300)]
[mir2loco] Set data format explicitly (#7419)

Pass `data_format` argument to conv-like constructors explicitly.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir] Fix padding computation (#7413)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Thu, 12 Sep 2019 13:45:29 +0000 (16:45 +0300)]
[mir] Fix padding computation (#7413)

Fix padding computation for `DeConv2DOp`.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[nnc] Pass data format to conv-like op constructors explicitly (#7409)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 14:18:45 +0000 (17:18 +0300)]
[nnc] Pass data format to conv-like op constructors explicitly (#7409)

* Pass data format to constructors of `Conv2DOp`, `DeConv2DOp`, `DepthwiseConv2DOp` explicitly.
* Add missing parameter to call of `DeConv2DOp` constructor.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[nnc] Unittests for Layout Switcher (#7321)
Павел Ильютченко/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 12:56:47 +0000 (15:56 +0300)]
[nnc] Unittests for Layout Switcher (#7321)

* Implemented tests for all layout related operations

Signed-off-by: Pavel Iliutchenko <p.iliutchenk@samsung.com>
5 years ago[mir] Remove old `DeConv2DOp` constructors (#7405)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 12:21:28 +0000 (15:21 +0300)]
[mir] Remove old `DeConv2DOp` constructors (#7405)

* Remove old `DeConv2DOp` constructors which accept only one padding vector instead of two.
* Add enum members to `PaddingType` to match ONNX specification.
* Adjust padding and output shape calculation for `DeConv2DOp`.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[exo-tflite] Make unit test use perm initialized with NHWC layout (#7396)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 08:58:00 +0000 (17:58 +0900)]
[exo-tflite] Make unit test use perm initialized with NHWC layout (#7396)

This make FeatureEncode and FeatureDecode use NHWC layout. Previous test code does not have this info and this test will fail when MaxPool2DConverter is enabled.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[exo-tflite] Introducing MaxPool2DConverter (#7387)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 08:20:08 +0000 (17:20 +0900)]
[exo-tflite] Introducing MaxPool2DConverter (#7387)

* [exo-tflite] Introducing MaxPool2DConverter

This adds MaxPool2DConverter, which converts loco::MaxPool2D to locoex::TFLMaxPool2D.

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

5 years ago[loco] Add TransposedConv2DLayer for test GraphBuilder (#7392)
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 08:14:33 +0000 (17:14 +0900)]
[loco] Add TransposedConv2DLayer for test GraphBuilder (#7392)

This commit adds TransposedConv2DLayer for test GraphBuilder.

Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
5 years ago[loco] Introduce Matrix Domain with its codecs (#7271)
Павел Ильютченко/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 08:12:43 +0000 (11:12 +0300)]
[loco] Introduce Matrix Domain with its codecs (#7271)

* Added Domain::Matrix
* Added MatrixShape, MatrixAxis, MatrixIndex
* Added and implemented MatrixEncoder and MatrixDecoder

Signed-off-by: Pavel Iliutchenko <p.iliutchenk@samsung.com>
5 years ago[mir2loco] Support DataFormat from mir in layout related ops (#7322)
Павел Ильютченко/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 08:08:55 +0000 (11:08 +0300)]
[mir2loco] Support DataFormat from mir in layout related ops (#7322)

* Created createHWIOFilterEncode
* Fixed createFeatureEncode, createFeatureDecode using DataFormat from
mir operations

Signed-off-by: Pavel Iliutchenko <p.iliutchenk@samsung.com>
5 years ago[ann-ref] Resolve SOLVe issues in ann-ref (#7380)
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 08:06:03 +0000 (17:06 +0900)]
[ann-ref] Resolve SOLVe issues in ann-ref (#7380)

This commit will add missing Tensorflow author and source statements

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
5 years ago[exo-tflite] Rename existing Pad class (#7388)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 08:05:36 +0000 (17:05 +0900)]
[exo-tflite] Rename existing Pad class (#7388)

This commit is for renaming existing Pad class in exo-tflite.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years ago[loco] Rename existing Pad class (#7393)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 08:04:55 +0000 (17:04 +0900)]
[loco] Rename existing Pad class (#7393)

This commit is for renaming existing Pad class in loco.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years ago[exo-tflite] Fix wrong ifdef guard (#7391)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 08:01:39 +0000 (17:01 +0900)]
[exo-tflite] Fix wrong ifdef guard (#7391)

This file was previously in `conversion` directory and ifdef guard contains CONVERSION. Now the file is in `src` so `CONVERSION` was removed.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[tfkit] Resolve SOLVe issues in tfkit (#7381)
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 07:58:05 +0000 (16:58 +0900)]
[tfkit] Resolve SOLVe issues in tfkit (#7381)

This commit will add Tensorflow author and remove redundant license statements

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
5 years ago[logo] Preparing SimplifyDomainConversionPass to add more visit() (#7157)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 07:56:52 +0000 (16:56 +0900)]
[logo] Preparing SimplifyDomainConversionPass to add more visit() (#7157)

* [logo] Preparing SimplifyDomainConversionPass to add more visit()

This is a preparation to introduce other visit(..).

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* nullifying input, etc

5 years ago[nnc] Resolve SOLVe issues in nnc (#7383)
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 07:56:07 +0000 (16:56 +0900)]
[nnc] Resolve SOLVe issues in nnc (#7383)

This commit will modify Apache 2.0 License to BSD 2-Clause License
since the code itself is 100% matched with original code and
caffe is under BSD 2-Clause License

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
5 years ago[mir2loco] Rename existing Pad class (#7386)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 07:51:16 +0000 (16:51 +0900)]
[mir2loco] Rename existing Pad class (#7386)

This commit is for renaming existing Pad class in mir2loco.

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years ago[mocotest-tf] Enable Some Broadcasting Tests (#7385)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 07:40:55 +0000 (16:40 +0900)]
[mocotest-tf] Enable Some Broadcasting Tests (#7385)

This commit enables supported broadcasting tests.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[nnpkg] Update specification for operators (#7355)
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 07:36:33 +0000 (16:36 +0900)]
[nnpkg] Update specification for operators (#7355)

It specifies the operators will use the same input, output and
attributes from tensorflow lite.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[moco-tf] Rename existing Pad class (#7384)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 07:17:20 +0000 (16:17 +0900)]
[moco-tf] Rename existing Pad class (#7384)

This commit is for renaming existing Pad class in moco-tf.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years ago[moco-tf] Support Broadcasting (#7378)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 07:09:36 +0000 (16:09 +0900)]
[moco-tf] Support Broadcasting (#7378)

This commit revises canonicalize_eltwise_binary_node to insert proper
boradcasting nodes if necessary.

This change will affect the canonicalization process of TFAdd, TFSub, TFMul, TFDiv.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[exo-tflite] exporting TFLMaxPool2D (#7382)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 07:09:23 +0000 (16:09 +0900)]
[exo-tflite] exporting TFLMaxPool2D (#7382)

Code to export TFLMaxPool2D was added. Since much of this code is duplicated with TFLAveragePool2D, common code was extracted into a separate method.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years agoAdd UNIT_Add_005 TensorFlow testcase (#7379)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 07:08:57 +0000 (16:08 +0900)]
Add UNIT_Add_005 TensorFlow testcase (#7379)

This commit adds UNIT_Add_005 TensorFlow testcase. This test is almost
same as UNIT_Add_001 except that it requires broadcasting over LHS (not RHS).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[neurun] Rename Scheduler to HEScheduler (#7340)
Дилшоджон Пошшоев/AI Tools Lab /SRR/Engineer/삼성전자 [Wed, 11 Sep 2019 07:08:36 +0000 (10:08 +0300)]
[neurun] Rename Scheduler to HEScheduler (#7340)

Since this is a scheduler for heterogeneous execution scheduling,
it must be called appropriately

Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
5 years ago[exo-tflite] Introduce TensorShapeExpander (#7373)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 07:02:06 +0000 (16:02 +0900)]
[exo-tflite] Introduce TensorShapeExpander (#7373)

* [exo-tflite] Introduce TensorShapeExpander

This will introduce TensorShapeExpander in TFLShapeInferenceRule for shape expansion for operations that has broadcasting

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* remove incs

5 years ago[loco] add alias before renaming pad class (#7377)
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 06:31:17 +0000 (15:31 +0900)]
[loco] add alias before renaming pad class (#7377)

This commit is getting ready for renaming pad class.

Related: #7357

Signed-off-by: seongwoo <sw4670.chae@samsung.com>
5 years agoAdd BSD-2-Clause license (#7370)
이춘석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 05:44:58 +0000 (14:44 +0900)]
Add BSD-2-Clause license (#7370)

- Add license in nnfw.spec
- Add license text

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
5 years ago[locomotiv] Simplify Element-wise Unary Op Kernels (#7358)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 05:34:54 +0000 (14:34 +0900)]
[locomotiv] Simplify Element-wise Unary Op Kernels (#7358)

* [locomotiv] Simplify Element-wise Unary Op Kernels

This commit rewrites element-wise unary op kernels on top of shared
infrastructure.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Fix copy-and-paste error

* Fix error message

5 years ago[exo-tflite] Adding ReluConverter test (#7287)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:33:50 +0000 (14:33 +0900)]
[exo-tflite] Adding ReluConverter test (#7287)

* [exo-tflite] Adding ReluConverter test

Two test cases were added: loco::relu with tensor input/putput, loco::relu with feature input/output.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* remove code that checks knob

* using plan.at(n) instead of bunch of  statements

* using find_first_node_bytype

* no loop, used iterator

5 years ago[exo-tflite] introducing knob for TFLMaxPool2D (#7376)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:32:47 +0000 (14:32 +0900)]
[exo-tflite] introducing knob for TFLMaxPool2D (#7376)

This adds knob for TFLMaxPool2D, for backward compatibility reason.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[moco-tf] Introduce CanonicalEltwiseInputConnector (#7368)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 05:31:25 +0000 (14:31 +0900)]
[moco-tf] Introduce CanonicalEltwiseInputConnector (#7368)

* [moco-tf] Introduce CanonicalEltwiseInputConnector

This commit introduces CanonicalEltwiseInputConnector which facilitates
pipeline extension.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Use NodeTy

5 years ago[exo-tflite] Adding AvgPool2DConverter (#7317)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:29:59 +0000 (14:29 +0900)]
[exo-tflite] Adding AvgPool2DConverter (#7317)

* [exo-tflite] Adding AvgPool2DConverter

AvgPool2DConverter is added, which converts loco::AvgPool2D to locoex::TFLAveragePool2D.

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

* fix comments

* fix typo

5 years ago[exo-tflite] TypeInference for TFLMaxPool2D (#7350)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:29:09 +0000 (14:29 +0900)]
[exo-tflite] TypeInference for TFLMaxPool2D (#7350)

This adds type inference for `TFLMaxPool2D`.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[exo-tflite] shape inference for TFLMaxPool2D (#7374)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:28:25 +0000 (14:28 +0900)]
[exo-tflite] shape inference for TFLMaxPool2D (#7374)

Adding shape inference for TFLMaxPool2D

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[logo] Revise test to use make_graph (#7371)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:23:33 +0000 (14:23 +0900)]
[logo] Revise test to use make_graph (#7371)

This will revise ConstantFoldingPass test to use make_graph() method instead of using stack value

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[locop] Do NOT show Shape attribute of Pull node (#7338)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 05:04:50 +0000 (14:04 +0900)]
[locop] Do NOT show Shape attribute of Pull node (#7338)

"shape" attribute in Pull node is deprecated.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[exo-tflite] shape inference for TFLAveragePool2D (#7345)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 05:04:37 +0000 (14:04 +0900)]
[exo-tflite] shape inference for TFLAveragePool2D (#7345)

Adding shape inference for TFLAveragePool2D and two test cases.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[moco-tf] Introduce Broadcast Helper (#7334)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 05:02:47 +0000 (14:02 +0900)]
[moco-tf] Introduce Broadcast Helper (#7334)

* [moco-tf] Introduce Broadcast Helper

This commit introduces Broadcast Helper module which facilitates
numpy-style broadcasting support.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Fix a typo

5 years ago[moco-tf] Remove unused CanonicalNodes.lst (#7366)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 03:52:11 +0000 (12:52 +0900)]
[moco-tf] Remove unused CanonicalNodes.lst (#7366)

This will remove not used anymore CanonicalNodes.lst file

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[moco-tf] Remove unused NodeShape (#7367)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 03:51:46 +0000 (12:51 +0900)]
[moco-tf] Remove unused NodeShape (#7367)

This will remove not used anymore moco::tf::NodeShape

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[exo-tflite] Convert for EltwiseAdd (#7353)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 03:34:32 +0000 (12:34 +0900)]
[exo-tflite] Convert for EltwiseAdd (#7353)

This will add converter for canonical EltwiseAdd to TFLAdd

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[exo-tflite] Type inference for TFLAdd (#7354)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 11 Sep 2019 03:34:05 +0000 (12:34 +0900)]
[exo-tflite] Type inference for TFLAdd (#7354)

This will add type inference for TFLAdd node

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[tfinfo] Rename targets (#7369)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 03:33:12 +0000 (12:33 +0900)]
[tfinfo] Rename targets (#7369)

* [tfinfo] Rename targets

This commit renames targets in tfinfo project to have tfinfo as prefix.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Fix a typo

5 years ago[moco-tf] Introduce canonicalize_eltwise_binary_node (#7327)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 11 Sep 2019 00:44:03 +0000 (09:44 +0900)]
[moco-tf] Introduce canonicalize_eltwise_binary_node (#7327)

* [moco-tf] Introduce canonicalize_eltwise_binary_node

A similar pattern is repeated across several element-wise binary
operation converters.

This commit extracts this pattern as canonicalize_eltwise_binary_node
to reduce code duplication.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Put implementations in anonymous namespace

5 years ago[loco] Remove unused member of FeatureEncodeLayer (#7344)
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 00:22:35 +0000 (09:22 +0900)]
[loco] Remove unused member of FeatureEncodeLayer (#7344)

This commit removes unused member '_perm' of FeatureEncodeLayer.

Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
5 years ago[loco] Add FilterEncodeLayer for test GraphBuilder (#7343)
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 11 Sep 2019 00:22:13 +0000 (09:22 +0900)]
[loco] Add FilterEncodeLayer for test GraphBuilder (#7343)

This commit adds FilterEncodeLayer for test GraphBuilder.

Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
5 years ago[mir] Switch to new `DeConv2DOp` constructor (#7361)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 13:57:28 +0000 (16:57 +0300)]
[mir] Switch to new `DeConv2DOp` constructor (#7361)

Construct `DeConv2DOp` with the new constructor.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir_caffe] Switch to new `DeConv2DOp` constructor (#7362)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 13:57:10 +0000 (16:57 +0300)]
[mir_caffe] Switch to new `DeConv2DOp` constructor (#7362)

Construct `DeConv2DOp` with the new constructor.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[mir] Add a new constructor to `DeConv2DOp` (#7359)
Сергей Баранников/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 13:22:24 +0000 (16:22 +0300)]
[mir] Add a new constructor to `DeConv2DOp` (#7359)

All existing constructors miss the "padding after" parameter. They will be replaced with the new one.

Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
5 years ago[cop, neurun] Use cpu backend by default for custom op (#7346)
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 09:54:17 +0000 (18:54 +0900)]
[cop, neurun] Use cpu backend by default for custom op (#7346)

`neurun` uses `acl_cl` backends by default.
However, for custom op, cpu backend will be used by default.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[exo-tflite] Modifying TODO in knob.lst (#7347)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 09:49:19 +0000 (18:49 +0900)]
[exo-tflite] Modifying TODO in knob.lst (#7347)

* [exo-tflite] Modifying TODO in knob.lst

Modifying names in TODO from TFL names to canonical names.

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

5 years agoImplement backend manipulation API (#7335)
이춘석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 09:36:59 +0000 (18:36 +0900)]
Implement backend manipulation API (#7335)

* Implement backend manipulation API

- Implement backend manipulation API

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* Apply suggestion

* fix format

5 years ago[nnpkg] update custom operator specification (#7337)
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 09:17:40 +0000 (18:17 +0900)]
[nnpkg] update custom operator specification (#7337)

* [nnpkg] update custom operator specification

It updates custom operator specification.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
* Update custom operator specificaiton

- Remove `nncc`
- Change the sentence as requested by review

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[neurun] Eliminate redundant fn in acl KernegGen (#7302)
Дилшоджон Пошшоев/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 08:54:48 +0000 (11:54 +0300)]
[neurun] Eliminate redundant fn in acl KernegGen (#7302)

No need to do extra move for most of the functions

Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
5 years ago[cop] Set isNnAPi=true for CustomNode (#7341)
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 08:49:07 +0000 (17:49 +0900)]
[cop] Set isNnAPi=true for CustomNode (#7341)

It sets isNnAPI for CustomNode to enable backend selection.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[locomotiv] Decouple NodeDataImpl and NodeAnnotation (#7332)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 08:30:32 +0000 (17:30 +0900)]
[locomotiv] Decouple NodeDataImpl and NodeAnnotation (#7332)

NodeDataImpl now inherits NodeData interface, which allows users to
annotate any NodeData implementation other than NodeDataImpl.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[exo-tflite] Adding TFLAveragePool2D into OperationExporter (#7315)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 07:00:41 +0000 (16:00 +0900)]
[exo-tflite] Adding TFLAveragePool2D into OperationExporter (#7315)

TFLAveragePool2D was added into OperationExporter.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years agoAPI to manipulate backend setting (#7298)
이춘석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 04:56:15 +0000 (13:56 +0900)]
API to manipulate backend setting (#7298)

Two functions(nnfw_set_default_backend, nnfw_set_op_backend) are
added.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
5 years agoFix error during building custom op app for pacl (#7329)
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 04:18:11 +0000 (13:18 +0900)]
Fix error during building custom op app for pacl (#7329)

Only nuerun supports custom op.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[custom op] Add example application (#7313)
Vladimir Plazun/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 02:10:45 +0000 (05:10 +0300)]
[custom op] Add example application (#7313)

* [custom op] Add example application

Add example application using FillFrom kernel and FillFrom nnpackage

Signed-off-by: Vladimir Plazun <v.plazun@samsung.com>
* fix wrong kernel name

5 years ago[custom op] Add example nnpackage (#7312)
Vladimir Plazun/AI Tools Lab /SRR/Engineer/삼성전자 [Tue, 10 Sep 2019 02:10:01 +0000 (05:10 +0300)]
[custom op] Add example nnpackage (#7312)

Adds example model using FillFrom custom operation

Signed-off-by: Vladimir Plazun <v.plazun@samsung.com>
5 years ago[moco-tf] Use shape_inference_done for fix_shape COpCall (#7325)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 01:47:25 +0000 (10:47 +0900)]
[moco-tf] Use shape_inference_done for fix_shape COpCall (#7325)

This will update fix_shape for COpCall to use shape_inference_done helper function

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[exo-tflite] Rename ConvertAdd (#7323)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 10 Sep 2019 01:46:16 +0000 (10:46 +0900)]
[exo-tflite] Rename ConvertAdd (#7323)

This will rename ConvertAdd to loco IR name ConvertEltwiseAdd

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[locomotiv] Support TensorBroadcast (#7309)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 01:34:41 +0000 (10:34 +0900)]
[locomotiv] Support TensorBroadcast (#7309)

* [locomotiv] Support TensorBroadcast

This commit extends locomotiv to support TensorBroadcast nodes.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Fix a copy-and-paste error

* Check input shape

5 years ago[loco] TransposedConv2D canonical IR (#7269)
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 10 Sep 2019 01:32:40 +0000 (10:32 +0900)]
[loco] TransposedConv2D canonical IR (#7269)

* [loco] TransposedConv2D canonical IR

This commit introduces TransposedConv2D, loco canonical IR for
spatial transposed convolution.

Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
* Reorder alphabetically

* Edit comment

5 years ago[loco] Refactor eltwise_binary_node_shape (#7296)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 00:03:58 +0000 (09:03 +0900)]
[loco] Refactor eltwise_binary_node_shape (#7296)

This commit relocates eltwise_binary_node_shape helper in order to
reduce the amount of code changes required for API v2 migration.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[loco] Do not invoke global shape_known helper (#7297)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 00:03:46 +0000 (09:03 +0900)]
[loco] Do not invoke global shape_known helper (#7297)

This commit revises ForwardShapeInferenceAlgorithm to invoke internal
"shape_known" helper instead of global shape_known helper.

This change reduces the amount of code changes for API v2 migration.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[bino] Remove bitwise-or operator overloading (#7300)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 10 Sep 2019 00:03:28 +0000 (09:03 +0900)]
[bino] Remove bitwise-or operator overloading (#7300)

Recently introduced fipe provides more general function pipe
implementation.

Let's use the function pipe implementation in fipe!

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[loco] Implement Type/Shape Inference for TensorBroadcast (#7306)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Sep 2019 23:57:28 +0000 (08:57 +0900)]
[loco] Implement Type/Shape Inference for TensorBroadcast (#7306)

This commit implements type/shape inference for TensorBroadcast nodes.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[exo-tflite] Moving Knob check of each converter into convert.cpp (#7303)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:39:18 +0000 (08:39 +0900)]
[exo-tflite] Moving Knob check of each converter into convert.cpp (#7303)

* [exo-tflite] Moving Knob check of each converter into convert.cpp

Knob check of each converter was moved into `convert.cpp`.

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

5 years ago[moco-tf] Delete moved to logo (#7311)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:38:56 +0000 (08:38 +0900)]
[moco-tf] Delete moved to logo (#7311)

This will delete transforms that has been moved to logo

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[exo-tflite] Putting all converts header file into one header file (#7305)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:38:45 +0000 (08:38 +0900)]
[exo-tflite] Putting all converts header file into one header file (#7305)

* [exo-tflite] Putting all converts header file into one header file

Now Conversion.h will include all converter's header file. This is to keep `convert.cpp` clean.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* move conversion/conversions.h to src

5 years ago[exo-tflite] adding `const` to TFLNodes.h (#7307)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:38:31 +0000 (08:38 +0900)]
[exo-tflite] adding `const` to TFLNodes.h (#7307)

This adds missing `const` to TFLNodes.h

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[exo-tflite] type inference of TFLAveragePool2D (#7314)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:34:45 +0000 (08:34 +0900)]
[exo-tflite] type inference of TFLAveragePool2D (#7314)

This commit adds type inference of TFLAveragePool2D.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[exo-tflite] change knob name for loco::AvgPool2D (#7316)
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 23:34:29 +0000 (08:34 +0900)]
[exo-tflite] change knob name for loco::AvgPool2D (#7316)

Previous name `ConvertAveragePool2D` followed TFL node name.
So, renamed to ConvertAvgPool2D just like converter's name follows canonical name.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[custom op] Add FillFrom kernel example (#7310)
Vladimir Plazun/AI Tools Lab /SRR/Engineer/삼성전자 [Mon, 9 Sep 2019 11:26:23 +0000 (14:26 +0300)]
[custom op] Add FillFrom kernel example (#7310)

Add example for kernel developers

Signed-off-by: Vladimir Plazun <v.plazun@samsung.com>
5 years ago[custom op] Add basic custom op example structure (#7308)
Vladimir Plazun/AI Tools Lab /SRR/Engineer/삼성전자 [Mon, 9 Sep 2019 10:34:18 +0000 (13:34 +0300)]
[custom op] Add basic custom op example structure (#7308)

* [custom op] Add basic custom op example structure

Add directories and CMakeLists for apps and kernels

Signed-off-by: Vladimir Plazun <v.plazun@samsung.com>
* use add_subdirectories

Signed-off-by: Vladimir Plazun <v.plazun@samsung.com>
5 years ago[cli] Use GTest_AddTest (#7286)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Sep 2019 10:18:11 +0000 (19:18 +0900)]
[cli] Use GTest_AddTest (#7286)

Let's simplify CMakeLists.txt a bit with GTest_AddTest helper.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
5 years ago[moco-tf] Apply shape_inference_done (#7304)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 10:14:25 +0000 (19:14 +0900)]
[moco-tf] Apply shape_inference_done (#7304)

This will apply to use shape_inference_done() for fix_shape nodes

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[logo] Unit test for ConstantFoldingPass (#7301)
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 9 Sep 2019 10:00:51 +0000 (19:00 +0900)]
[logo] Unit test for ConstantFoldingPass (#7301)

* [logo] Unit test for ConstantFoldingPass

This will add unit test for ConstantFoldingPass copied from moco-tf ConstantFoldingTransform excluing import and log

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* apply comments

5 years ago[backend/srcn] Add transpose conv kernel (#7248)
윤지영/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Sep 2019 09:53:30 +0000 (18:53 +0900)]
[backend/srcn] Add transpose conv kernel (#7248)

* [backend/srcn] Add transpose conv kernel

Srcn library supports raw-major(NCHW) and col-major(NHWC) image layout.
However this commit only supports col-major image layout.

And Srcn library takes HWOI kernel filter layout when the image layout is NHWC.
But in neurun, it takes OHWI kernel filter layout when using the same image layout.
So it need to permute codes for TransposeConv weight.

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Change the Permute function to be limited to kernel layout

* Use original data array without copying to new address

5 years ago[locomotiv] Use const if possible (#7264)
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 9 Sep 2019 09:37:11 +0000 (18:37 +0900)]
[locomotiv] Use const if possible (#7264)

"annot_data" and "annot_domain" do not change the node itself.

This commit revies the signature of these helpers to express this
property explicitly.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>