박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 07:25:34 +0000 (16:25 +0900)]
[moco-tf] Remove unused WindowData header (#8121)
This will remove not used anymore WindowData.h file
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:55:34 +0000 (15:55 +0900)]
[moco-tf] Remove unused PadData and StrideData (#8115)
This will remove not used anymore PadData and StrideData annotations
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 14 Oct 2019 06:52:36 +0000 (15:52 +0900)]
[moco-tf] Turn ON knobs for resolving Reshape (#8117)
This commit turns on following knobs:
- ResolveDuplicateReshape
- ResolveRedundantReshape
Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 14 Oct 2019 06:49:58 +0000 (15:49 +0900)]
[moco-tf] Fix Placeholder import (#8114)
Previously, Placeholder with negative value was not caught by assertion,
and converted to Pull with overflowed value. This commit fixes this
behaviour
Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:49:37 +0000 (15:49 +0900)]
[tf2tflite] use exo (#8109)
This will update tf2tflite to use exo instead of exo_tflite
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:49:14 +0000 (15:49 +0900)]
[exo] Update includes (#8105)
This will remove not needed anymore header includes for ExportUtils
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:48:38 +0000 (15:48 +0900)]
[exo] use circle for comments and methods (#8111)
This will update to use Circle for comments and local methods
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 06:36:33 +0000 (15:36 +0900)]
[nnfw-api] Introduce resize input API (#8034)
* [nnfw-api] Introduce resize input API
Introduce API for resize input: nnfw_apply_tensorinfo
Internal implementation always return error - NYI
Update related API's comment based on doxygen format
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* update comment
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 06:24:14 +0000 (15:24 +0900)]
[nnapi-test] Use android 10 test generation (#8040)
Use android 10 test generation
- Disable TestCompliance util
- Disable _Float16 feature
- Disable dynamic output shape test generation
- Fix to use c++11 feature
- Update generated code archive for tizen build
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:13:34 +0000 (15:13 +0900)]
[nnfw-api] explain nnfw_set_backend and nnfw_set_op_backend more (#8071)
It explains more about nnfw_set_backend and nnfw_set_op_backend.
Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 06:12:10 +0000 (15:12 +0900)]
[nnpkg-tcs] Run nnpkg-tcs after tc exists (#8119)
Some nnpkg-tcs are not generated via `tf2tflite` test filter.
Thus, nnpkg-tcs checks whether test case exists before running.
Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 04:47:10 +0000 (13:47 +0900)]
[exo] Adding FuseConv2DAddSubPass into TFLOptimize.cpp (#8102)
Now FuseConv2DAddSubPass runs when UseFuseConv2DAddSubPass knob is on.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 04:43:20 +0000 (13:43 +0900)]
Let locomotiv be installable (#8089)
From now on, install command will put locomotive library into the proper
location.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 04:42:52 +0000 (13:42 +0900)]
[moco-tf] Rename TestHelper filename (#8104)
This will rename TestHelper to TestHelper.test so that it only exist for test executable not shared library
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 14 Oct 2019 04:41:03 +0000 (13:41 +0900)]
[moco-integration-test] Add renamed test case (#8098)
This commit adds renamed test case.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 04:34:56 +0000 (13:34 +0900)]
Let make moco-tf installable (#8093)
From now on, install command will put moco-tf library into a proper
location.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 04:34:43 +0000 (13:34 +0900)]
[exo] add copy of TFExporter as CircleExporter (#8094)
This will add a copy of TFExporter and related files as CircleExporter to export Circle file
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 04:25:41 +0000 (13:25 +0900)]
Let make tf2tflite installable (#8096)
From now on, install command will put tf2tflite binary into a proper
location.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 04:20:13 +0000 (13:20 +0900)]
[moco-tf] Remove ShapeInferenceData annotation (#8100)
This will remove ShapeInferenceData annotation that was replaced by loco NodeShape
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 03:46:06 +0000 (12:46 +0900)]
[moco-tf] TFShapeInferenceRule test (#8074)
This will add TFShapeInferenceRule test from old FixShapeTransform test
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 03:35:40 +0000 (12:35 +0900)]
[tflite2circle] Remove the limitation of running under nnfw root (#8068)
Now, it uses the pathname of the currently executing script to calculate the default path for other resources.
Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 14 Oct 2019 03:17:21 +0000 (12:17 +0900)]
[tf2tflite] Add renamed test case (#8088)
This commit adds renamed test case to tf2tflite.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 01:57:25 +0000 (10:57 +0900)]
[exo] Change target to exo (#8073)
* [exo] Change target to exo
This will update CMake target to exo
- add a temporary exo_tflite target for dependent targets
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* one more change
* add a dup instead of interface
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 14 Oct 2019 01:52:06 +0000 (10:52 +0900)]
[moco-tf] Add Pad canonicalizer (#8030)
* [moco-tf] Add Pad canonicalizer
This commit adds Pad canonicalizer to moco-tf.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* apply commnets.
* apply comment.
* Simplify method to get rank of padding.
* fix wrong assert.
* add comments in canonicalization.
* make comment more clear.
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 01:48:19 +0000 (10:48 +0900)]
[exo] fix namespace for local method (#8087)
This will fix namespace to anonymous for local method encodeOperatorCodes
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 01:33:15 +0000 (10:33 +0900)]
[moco] Add Pad IR (#8090)
This will copy Pad IR from moco-tf
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
[neurun] Implement circle_loader (#8079)
Implement `CircleLoader` by deriving it from `BaseLoader`.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 14 Oct 2019 00:45:33 +0000 (09:45 +0900)]
[nncc] Set proper RPATH on installation (#8057)
This commit makes install command (such as make install) to generate
a portable package.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 14 Oct 2019 00:39:00 +0000 (09:39 +0900)]
[moco-tf] Remove FixShapeTransform (#8075)
This will remove FixShapeTransform and related codes that is replaced by TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 10:38:52 +0000 (19:38 +0900)]
[cmake] Fix endif mismatching warning (#8070)
Fix if-endif mismatching warning in cker cmake
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 08:27:39 +0000 (17:27 +0900)]
[exo] Relocate and rename exo-tflite to exo (#8069)
This will relocate and rename exo-tflite to exo to prepare export of Circle format
- all tflite related files are moved inside TFLite folder
- local files have tflite_detail sub namespace
- some unused headers exist and will be removed in next PR
- CMake target is still exo_tflite and will be changed in next PR
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 08:22:58 +0000 (17:22 +0900)]
[cmd] Format checker check unstaged files and ignore untracked files (#8058)
Ignore untracted file on format checking command
Add modified file on develop branch format checking command
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 07:57:04 +0000 (16:57 +0900)]
[cker] cker optimized build option (#8065)
Introduce cker optimized build option to disable optimized cker build on eigen usable environment
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 07:56:45 +0000 (16:56 +0900)]
[loco] Install shared library (#8051)
TNow, it is possible to install "libloco.so" with install command (e.g. make install).
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 07:56:34 +0000 (16:56 +0900)]
[locoex-customop] Support Install (#8055)
Let's make it possible to install locoex-customop with install command.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 07:56:21 +0000 (16:56 +0900)]
[moco-log] Support Install (#8056)
Let's make it possible to install moco-log with install command!
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 06:59:11 +0000 (15:59 +0900)]
[exo-tflite] Test to check if transpose is exported OK (#8000)
* [exo-tflite] Test to check if transpose is exported OK
This adds two tests to see if transpose is exported OK.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* knob check
* revise comment for perm vector; removed unused vars
* add todo; replace 'if' to 'ASSERT_EQ'
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 06:55:48 +0000 (15:55 +0900)]
[moco-tf] Revise shape inf for TFShape and rest (#8053)
This will revise shape inference for TFShape and reset of the nodes to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 06:53:55 +0000 (15:53 +0900)]
[exo-tflite] Renaming DefaultLayout to FeatureLayout (#8046)
* [exo-tflite] Renaming DefaultLayout to FeatureLayout
DefaultLayout is renamed to to FeatureLayout.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* fix comment
박천교/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Oct 2019 06:33:07 +0000 (15:33 +0900)]
[logo] Constant folding skips TensorBroadcast (#7847)
* [logo] Constant folding skips TensorBroadcast
TensorBroadcast is a good hint for optimizaiton. Let's skip this for
constant folding.
Signed-off-by: Cheongyo Bahk <ch.bahk@samsung.com>
* Add TODO comment
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 06:30:49 +0000 (15:30 +0900)]
[exo-tflite] Code to perform fusing TFLAdd/TFLSub into TFLConv2D (#8037)
* [exo-tflite] Code to perform fusing TFLAdd/TFLSub into TFLConv2D
This adds code that fuses TFLAdd/TFLSub into TFLConv2D.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* remove unused param
* todo for cyclomatic complexity
김용섭/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Oct 2019 04:20:19 +0000 (13:20 +0900)]
[neurun] Prepare tryDeallocConst to ExecutorFactory (#7333)
* [neurun] Prepare tryDeallocConst to ExecutorFactory
Prepare tryDeallocConst to ExecutorFactory and remove unused lines
Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Add TODO
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Oct 2019 04:10:05 +0000 (13:10 +0900)]
[res] Add simple Pad network with constant (#7982)
* [res] Add simple Pad network with constant
This commit add simple Pad network with constant_value.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* rename test folder.
* rename menifest file.
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 11 Oct 2019 01:39:25 +0000 (10:39 +0900)]
Introduce tf2circle (#8038)
This commit introduces tf2circle with a dummy implementation in order to
create a room for co-work.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 01:26:06 +0000 (10:26 +0900)]
[res/TensorFlowTests] Remove "use_cudnn_on_gpu" attrib in pbtxt files (#8041)
"use_cudnn_on_gpu" was removed from all pbtxt files in 'res/TensorFlowTests'.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Fri, 11 Oct 2019 01:04:16 +0000 (10:04 +0900)]
[moco-tf] Revise shape inf for Mean Mul (#8017)
* [moco-tf] Revise shape inf for Mean Mul
This will revise shape inference for TFMean and TFMul node to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* remove unused var
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 23:37:36 +0000 (08:37 +0900)]
[moco-tf] Add Pad GraphBuilder (#8032)
* [moco-tf] Add Pad GraphBuilder
This commit adds Pad GraphBuilder to moco-tf.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* check whether it has a right attribute.
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 23:37:08 +0000 (08:37 +0900)]
[exo-tflite] Rewriting a test using TestGraph (#8045)
minimal_with_TFLRelu is re-written with TestGraph to shorten code.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 23:31:45 +0000 (08:31 +0900)]
[exo-tflite] Introducing TFLRelu6 (#8048)
This introduce TFLRelu6 node, which is one of ops that can be fused into conv2d.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 23:31:19 +0000 (08:31 +0900)]
[exo-tflite] Fix wrong logging input name of Relu (#8049)
TFLRelu has `features()` as its input. This commit fix wrong name ("input") to "features".
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 22:04:41 +0000 (07:04 +0900)]
[tf2tflite] update exo-tflite dependency (#8035)
This will update tf2tflite to use exo to prepare exporter migration to exo folder
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 10:14:14 +0000 (19:14 +0900)]
Change model IR for ResizeBilinear param (#8025)
This commit changes model IR for ResizeBilinear param to scalar.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Oct 2019 10:06:12 +0000 (19:06 +0900)]
[nnfw-cmake] Support gmock library (#8031)
Prepare to support gmock - android 10 nnapi test require gmock
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 09:49:27 +0000 (18:49 +0900)]
[res] Add Pad network with REFLECT mode (#7984)
* [res] Add Pad network with REFLECT mode
This commit add Pad network with REFLECT mode.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* fix info.
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 09:43:45 +0000 (18:43 +0900)]
[moco-tf] Revise shape inf for RealDiv and others (#8021)
This will revise shape inference for TFRealDiv and other nodes starting with R to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 09:22:42 +0000 (18:22 +0900)]
[exo-tflite] turning a knob : ConvertTensorTranspose (#8033)
ConvertTensorTranspose knob became true by default.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 09:22:03 +0000 (18:22 +0900)]
[exo-tflite] Introducing a pass that fuse TFLConv2D and [TFLAdd or TFLSub] (#7995)
* [exo-tflite] Introducing a pass that fuse TFLConv2D and [ TFLAdd or TFLSub ]
This introduces a pass that fuse TFLConv2D and [ TFLAdd or TFLSub ]. Real fusing code will be committed in next PR.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* renaming new_bias_node to widened_const
* removed comment that says bias of conv2d is all 0's
* fix type + use more specific node names
* stop using loco::replace(..) for widened bias const
* removed unused vars
* bug fix for set_const_input
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 08:46:37 +0000 (17:46 +0900)]
[exo-tflite] turning on some knobs (#8002)
This turns knobs of some knobs. (all knobs except knobs related to Conv2D)
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 08:46:01 +0000 (17:46 +0900)]
[res] Rename testcase as convention (#8029)
* [res] rename testcase as convention
This commit renames some testcase as naming convention.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* apply comment.
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 08:38:37 +0000 (17:38 +0900)]
[res] Test case to check fusing [ Conv2D - Sub ] (#7991)
Adding a test case to check fusing [ conv2d - sub ] subnetwork into Conv2D of which bias is set to y value of Sub.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 08:34:06 +0000 (17:34 +0900)]
[exo] Empty exo project for migration (#8028)
This will add empty exo project for migration of exo-tflite and to support circle files
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 08:31:42 +0000 (17:31 +0900)]
[android] Enable TF Lite JNI build (#8026)
Enable TF Lite JNI module build which will be used by Android Java Apps.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 04:32:39 +0000 (13:32 +0900)]
[nnpkg-run] set default backend using OP_BACKEND_ALLOPS environment var. (#8003)
`nnpackage-run` (in short, `nnpkg-run`) uses `nnfw_session` from nnfw APIs.
It uses GeneralConfigSource. It means we need to set configuration using nnfw APIs.
Thus, it reads from environment var and set default backends.
Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
여지환/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Oct 2019 03:58:32 +0000 (12:58 +0900)]
Update document for docker image creation (#8016)
Build command changed from 'nnfw' to 'nnas'.
Update https://github.sec.samsung.net/STAR/nnfw/blob/master/docs/nnfw/howto/HowToUseDockerImage.md to use 'nnas' instead of 'nnfw'.
Signed-off-by: JiHwan Yeo <jihwan.yeo@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Oct 2019 03:56:40 +0000 (12:56 +0900)]
[nnapi-test] Port TestMain (#7907)
Port TestMain.cpp from upstream android nn runtime's NNAPI test
Fix
- Always define NNTEST_ONLY_PUBLIC_API to avoid android dependency
- Comment out android logging related code
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 01:52:31 +0000 (10:52 +0900)]
[moco-tf] Revise shape inf for FusedBatchNorm and Identity (#8014)
This will revise shape inference for TFFusedBatchNorm and TFIdentity node to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Oct 2019 01:48:28 +0000 (10:48 +0900)]
[loco] Introduce Pad IR (#7992)
This commit introduces Pad IR to loco.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
[neurun] Templatize BaseLoader (#8012)
* Templatize `BaseLoader` class.
* Add tflite-specific `LoaderDomain` structure.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Thu, 10 Oct 2019 00:05:29 +0000 (09:05 +0900)]
[moco-tf] Revise shape inf for Const, Conv2D and Conv2DBackpropInput (#7999)
This will revise shape inference for TFConst, TFConv2D and TFConv2DBackpropInput node to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Wed, 9 Oct 2019 22:07:12 +0000 (07:07 +0900)]
[moco-tf] Introduce node_shape_with_check helper (#7993)
This will introduce node_shape_with_check() helper in TFShapeInferenceRule for getting node_shape of a node and revise for TFBiasAdd node
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
[adtidas] Reformat using clang-format (#8013)
* Reformat the component using clang-format.
* Enable format checking.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
[mir_tflite] Change the stream type to char (#8006)
`uint8_t` is not supported directly by the standard library, leading to errors in runtime.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
Enable regression test for tf2tflite utility (#8005)
- Enabled Regression test 2 that checks tf2tfite correctly handles conv2d withour dilation
- Enabled Regression test 1
Signed-off-by: Efimov Alexander <a.efimov@samsung.com>
Enable regression test for moco importer (#8004)
- Enabled regression test 2 that checks that tf-moco correctly handles conv2d withour dilation
- Enabled Regression test 1
Signed-off-by: Efimov Alexander <a.efimov@samsung.com>
김용섭/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 11:04:17 +0000 (20:04 +0900)]
[neurun] Eliminate nnapi tests have different spec (#7290)
Eliminate nnapi tests which have different spec from acl.
nnapi can take weight/bias as input but acl can't.
Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
Regression test for non dilated conv2d (#7970)
Test to check that convolution without dilation attribute compiles successfully
Signed-off-by: Efimov Alexander <a.efimov@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 09:31:21 +0000 (18:31 +0900)]
[moco-tf] Revise shape inference for MaxPool (#7996)
This will revise shape inference for MaxPool to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 09:12:47 +0000 (18:12 +0900)]
[exo-tflite] Fix to use ActivationFunction (#7983)
Until now, `exo-tflite` didn't use `ActivationFunction` when making an options
even if some operations have it.
This commit will fix it to use ActivationFunction
Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 08:59:14 +0000 (17:59 +0900)]
[moco] Revise AvgPool Conv2D input names (#7989)
This will fix AvgPool and Conv2D input name to follow that from TensorFlow REGISTER_OP input names
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 07:51:56 +0000 (16:51 +0900)]
[moco] Revise TFFusedBatchNorm input (#7975)
* [moco] Revise TFFusedBatchNorm input
This will revise TFFusedBatchNorm IR input name to follow TensorFlow document
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* use x for input
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 07:51:37 +0000 (16:51 +0900)]
[moco-tf] Introduce TFPad IR (#7959)
* [moco-tf] Introduce TFPad IR
This commit introduces TFPad IR to moco-tf.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* apply comments.
* move location of function declaration alphabetical order.
* rename input method as convention and remove unnecessary assert.
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 07:46:13 +0000 (16:46 +0900)]
[exo-tflite] Introduce 'optimize(..)' that optimize TFL Nodes (#7958)
* [exo-tflite] Introduce 'optimize(..)' that optimize TFL Nodes
This adds 'optimize(..)' that optimize TFL Nodes.
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
* optimize.cpp to TFLOptimize.cpp
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 8 Oct 2019 07:44:19 +0000 (16:44 +0900)]
[cmd] Docker image build command (#7981)
Add argument in build docker image command for additional dockerfile - ubuntu 18.04
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 07:21:38 +0000 (16:21 +0900)]
[loco] Introduce PaddingND (#7949)
* [loco] Introduce PaddingND
This commit introduces PaddingND class to loco for supporting Pad.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
* rename size method to rank.
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 07:19:28 +0000 (16:19 +0900)]
[moco-tf] ShapeInf rev for DepthwiseConv2dNative (#7974)
This will revise shape inference for DepthwiseConv2dNative to be done in TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 07:07:31 +0000 (16:07 +0900)]
Make to support BatchToSpaceND op for acl cl (#7940)
This commit makes to support BatchToSpaceND op for acl cl.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 07:04:37 +0000 (16:04 +0900)]
Remove layout from OperandInfo (#7962)
This commit removes layout from OperandInfo.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 05:06:39 +0000 (14:06 +0900)]
Update Makefile.template for Android (#7919)
Pass and set additional CMake variables for Android build.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 04:56:38 +0000 (13:56 +0900)]
Introduce Dockerfile based on Ubuntu 18.04 (#7973)
This is an identical file with `infra/docker/Dockerfile` except for
base Ubuntu version. This will be used for Android build which requires
higher version of CMake.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
채성우/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 03:17:34 +0000 (12:17 +0900)]
[loco] Rename pad.test.cpp file (#7952)
This commit renames pad.test.cpp file to Padding2D.test.cpp.
Signed-off-by: seongwoo <sw4670.chae@samsung.com>
[neurun] Make some functions members of `BaseLoader` class (#7966)
* Move standalone functions `convertActivation`, `tensorTypeToDataType` and `lopadOperationIO` inside `BaseLoader` class.
* Rewrite `convertActivation` to use `switch` instead of `static_cast`.
* Improve some error messages a little.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 01:56:11 +0000 (10:56 +0900)]
[res] Add testcases for merging concat (#7961)
This commit will add testcases for `MergeConcat` optimization
Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
남궁석/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Oct 2019 01:55:12 +0000 (10:55 +0900)]
[exo-tflite] Introduce locoex::FusedActFunc converter (#7964)
This commit will introduce `locoex::FusedActFunc` converter,
to return `tflite::ActivationFunctionType`
Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 01:24:29 +0000 (10:24 +0900)]
[moco-tf] ShapeInf for TFConcatV2 (#7947)
* [moco-tf] ShapeInf for TFConcatV2
This will update shape inference to be done in TFShapeInferenceRule for TFConcatV2 node and not use ConcatData annotation
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
* apply comments
* simplify assert check
윤현식/On-Device Lab(SR)/Principal Engineer/삼성전자 [Tue, 8 Oct 2019 01:10:46 +0000 (10:10 +0900)]
[exo-tflite] exporting tflite CONV_2D op by using TFLConv2d (#7941)
This code enables tf2tflite to export tflite CONV_2D op by using TFLConv2d. (when knob is on)
Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
[mir2loco] Fix Pad/Window/Stride initialization (#7968)
The order of height / width was messed up.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
[mir_tflite] Change the type of `_tensorMap` to `vector`. (#7967)
`vector` is more efficient and less space consuming.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>
박세희/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 7 Oct 2019 10:30:28 +0000 (19:30 +0900)]
[moco_tf] Remove FixShapeTransform.test.cpp (#7955)
This will remove FixShapeTransform.test.cpp file
- FixShapeTransform will be replaced by TFShapeInferenceRule
Signed-off-by: SaeHie Park <saehie.park@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Mon, 7 Oct 2019 10:17:21 +0000 (19:17 +0900)]
Make to support SpaceToBatchND op for acl neon (#7950)
This commit make to support SpaceToBatchND op for acl neon.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
[neurun] Introduce BaseLoader (#7914)
Rename `TFLiteLoader` class to `BaseLoader` and move it from `tflite/tflite_loader.cc` to `base_loader/base_loader.h`.
Signed-off-by: Sergei Barannikov <s.barannikov@samsung.com>