platform/core/ml/nnfw.git
5 years agoFix padding bug of some kernels. (#3161)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 06:14:46 +0000 (15:14 +0900)]
Fix padding bug of some kernels. (#3161)

This commit Fixes padding bug of some kernels used by Coordinates of ACL.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years ago[neurun] Apply BumpAllocator to TensorBuilder (#3172)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 06:12:50 +0000 (15:12 +0900)]
[neurun] Apply BumpAllocator to TensorBuilder (#3172)

Applies BumpAllocator to TensorBuilder so that now one memory allocator
has started allocating tensors.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
5 years agoversion update to 0.2 on gbs spec file (#2735) (#3156)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 06:07:54 +0000 (15:07 +0900)]
version update to 0.2 on gbs spec file (#2735) (#3156)

- version update to 0.2 on gbs spec file

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
5 years agoAdd a generated test for testing broadcast SUB (#3144)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 06:00:40 +0000 (11:30 +0530)]
Add a generated test for testing broadcast SUB (#3144)

This test is for testing broadcast sub with tensors of 4D & 2D shape.
Added GTest: sub_broadcast_4D_2D_float_nnfw.
Related issue: #3143

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoFix CRLF to LF in Abs.cpp (#3175)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 16 Oct 2018 05:57:33 +0000 (14:57 +0900)]
Fix CRLF to LF in Abs.cpp (#3175)

Fix wrong line break in Abs.cpp.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
5 years ago[cherry-pick] Remove debug files in package (#2778) (#3158)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 04:37:53 +0000 (13:37 +0900)]
[cherry-pick] Remove debug files in package (#2778) (#3158)

Remove debug files in nnfw & nnfw-test package

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years ago[cherry-pick] Remove unused module on gbs build (#2755) (#3157)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 04:11:18 +0000 (13:11 +0900)]
[cherry-pick] Remove unused module on gbs build (#2755) (#3157)

Remove unused module on gbs build

- tools (except nnapi_test)
- benchmark
- contrib
- template & logging in runtime

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years agotflite_benchmark custom op on tizen (#3170)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 01:53:17 +0000 (10:53 +0900)]
tflite_benchmark custom op on tizen (#3170)

On tizen, custom op is not found when runnning tflite_benchmark.
This patch changes link order for tflite_benchmark.
cf) This bug is not found on arm-ubuntu(IDK why)

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
5 years agoAdd a generated test for broadcast add (#3125)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:50:46 +0000 (10:50 +0900)]
Add a generated test for broadcast add (#3125)

This test is consist of an ADD, a CONCATENATION and a RESHAPE operation.
CONCATENATION and RESHAPE are NOP just copy the value to the output.
They are exist for making test easier with different layouts(NCHW/NHWC)
on neurun.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years ago[tfltool] OP selector: Support more operators (#3096)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:49:47 +0000 (10:49 +0900)]
[tfltool] OP selector: Support more operators (#3096)

Supports BATCH_TO_SPACE_ND and SPACE_TO_BATCH_ND operators

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
5 years ago[neurun] Introduce BumpAllocator (#3149)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:43:11 +0000 (10:43 +0900)]
[neurun] Introduce BumpAllocator (#3149)

* [neurun] Introduce BumpAllocator

Introduces BumpAllocator an early version of MemoryAllocator.
The BumpAllocator only allocates memory without any deallocating.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
* Adds unit test of BumpAllocator

* Remove capacity()

5 years ago[tfltool] OP selection: Add support for L2Norm and Split (#3109)
PRAVEEN DORESWAMY NAIDU/System SW /SRI-Bangalore/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 01:42:36 +0000 (07:12 +0530)]
[tfltool] OP selection: Add support for L2Norm and Split (#3109)

This patch will add support for L2Norm and Split operators
in tfltool select_operator.
Signed-off-by: Praveen D N <praveen.dn@samsung.com>
5 years agonnfw: RSQRT PACL Changes for RSQRT Support (#3040)
TANUJ TEKRIWAL/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:40:49 +0000 (07:10 +0530)]
nnfw: RSQRT PACL Changes for RSQRT Support (#3040)

This patch adds support for RSQRT in PACL Codebase(compilation.cc)

Signed-off-by: Tanuj Tekriwal <tanuj.tekri@samsung.com>
5 years agoRemove old ACL temp files from gitignore (#3115)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 16 Oct 2018 01:34:45 +0000 (10:34 +0900)]
Remove old ACL temp files from gitignore (#3115)

This will remove ACL generated files that don't need anymore in gitignore
- /externals/acl itself is ignored now

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[tftool] OP selector: Support TransposeConv op (#3133)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:32:51 +0000 (07:02 +0530)]
[tftool] OP selector: Support TransposeConv op (#3133)

For issue #3093
This patch will add support for transposeconv in tfltool selec_operator.py

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years ago[tfltool] OP selector: Support more operator (#3094)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:32:34 +0000 (10:32 +0900)]
[tfltool] OP selector: Support more operator (#3094)

This commit supports SQUEEZE option in op selector

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
5 years ago[neurun] Fix operation::Concat input assertion (#3164)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:28:24 +0000 (10:28 +0900)]
[neurun] Fix operation::Concat input assertion (#3164)

Just like the comment says, it should be equal or greater than 2.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoRemove passed GTest from skiplist (#3165)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:26:18 +0000 (06:56 +0530)]
Remove passed GTest from skiplist (#3165)

This patch removes sub_broadcast_float GTest from skiplist.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[tfltool] OP selection: Enable for EqualOptions (#3112)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:25:41 +0000 (06:55 +0530)]
[tfltool] OP selection: Enable for EqualOptions (#3112)

This patch will enable EqualOptions in tfltool select_operator.
Related issue: #2610 #2822

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoRemoving function overhead of pow in Squarediff kernel (#3130)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:23:26 +0000 (06:53 +0530)]
Removing function overhead of pow in Squarediff kernel (#3130)

This patch will remove function overhead of pow since pow(x,2) can be wirtten as x*x

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoCorrection in reduction mean kernel (#3131)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:18:52 +0000 (06:48 +0530)]
Correction in reduction mean kernel (#3131)

This patch removes unnecessary temp varaible and modifies comments

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoOffset function not required (#3132)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Tue, 16 Oct 2018 01:15:22 +0000 (06:45 +0530)]
Offset function not required (#3132)

This patch will remove unused offset4D function from CLReduceSum.cpp

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoUse dlopen for generated test (#3154)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Tue, 16 Oct 2018 01:14:33 +0000 (10:14 +0900)]
Use dlopen for generated test (#3154)

Use dlopen instead of static link for generated test

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years agoRemove duplicated codes from SimpleEmbeddingLookup (#3159)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 16 Oct 2018 01:08:57 +0000 (10:08 +0900)]
Remove duplicated codes from SimpleEmbeddingLookup (#3159)

This commit removes duplicated codes from SimpleEmbeddingLookup.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years agoAdd CRLF detect in format-checker (#3162)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 16 Oct 2018 00:33:55 +0000 (09:33 +0900)]
Add CRLF detect in format-checker (#3162)

This will add CRLF newline check in format-checker script

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years agoInitialize Objects and other variables (#3129)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Mon, 15 Oct 2018 11:09:34 +0000 (16:39 +0530)]
Initialize Objects and other variables (#3129)

This patch will initialize class objects and variables

Signed-off-by: Shubham <shub98.gupta@samsung.com>
5 years agoRemove unnecessary Headerfile <iostream> from op BatchToSpaceNd (#3108)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Mon, 15 Oct 2018 09:42:47 +0000 (15:12 +0530)]
Remove unnecessary Headerfile <iostream> from op BatchToSpaceNd (#3108)

This patch removes unnecessary headerfile <iostream> from op/BatchToSpaceNd.cc

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoAdd a function that converts the coordinates of the runtime into the coordinates...
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 15 Oct 2018 09:41:30 +0000 (18:41 +0900)]
Add a function that converts the coordinates of the runtime into the coordinates of ACL. (#3117)

This commit adds a function that converts the coordinates of the runtime into the coordinates of ACL.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years agoConvert CRLF line endings to LF (#3148)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 15 Oct 2018 09:40:55 +0000 (18:40 +0900)]
Convert CRLF line endings to LF (#3148)

Some files use CRLF(Windows) for line ending while most files use LF
(Unix-like).

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoRevert "[neurun] Only one item for codegen operand Context (#3043)" (#3127)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 15 Oct 2018 04:28:48 +0000 (13:28 +0900)]
Revert "[neurun] Only one item for codegen operand Context (#3043)" (#3127)

This reverts commit f607de26dcf88a20de5e1d303390f9253cb27c34.

There is no easy way to fix #3126, so reverting the problematic change.
However still, we need to fix it.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years ago[neurun] Remove unnecessary header includes (#3142)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 15 Oct 2018 03:11:31 +0000 (12:11 +0900)]
[neurun] Remove unnecessary header includes (#3142)

Remove more unnecessary header includes in
`frontend/wrapper/compilation.cc`.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoAdd Abs custom op in tflite (#3033)
Siva Sai Vaddipati/System SW /SRI-Bangalore/Engineer/삼성전자 [Mon, 15 Oct 2018 02:17:29 +0000 (07:47 +0530)]
Add Abs custom op in tflite (#3033)

This commit resgiters and performs Custom(Abs) op in tflite.
Related issue:#2828

Signed-off-by: Siva Sai <siva.sai@samsung.com>
5 years agoBlock size assert for space_to_depth (#3087)
PRAVEEN DORESWAMY NAIDU/System SW /SRI-Bangalore/Staff Engineer/삼성전자 [Fri, 12 Oct 2018 09:45:50 +0000 (15:15 +0530)]
Block size assert for space_to_depth (#3087)

The block size should be asserted for size, divisor of
height and weight and output depth as per NNAPI specification.
The change asserts as per specification.

Assert moved from execution to compile time.

Signed-off-by: Praveen D N <praveen.dn@samsung.com>
5 years agoRemove Unsupported Datatye Cases in StridedSlice (#3111)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 12 Oct 2018 09:45:27 +0000 (15:15 +0530)]
Remove Unsupported Datatye Cases in StridedSlice (#3111)

NN api defines only for U8 and F32 dataype for stridedslice op.
Hence, leaving code only for supported dataypes.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoAdd missing definition for Cast.h (#3120)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 08:04:28 +0000 (17:04 +0900)]
Add missing definition for Cast.h (#3120)

This commit Add missing definition for Cast.h

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years agoChange external source code download setting (#3116)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 12 Oct 2018 07:51:12 +0000 (16:51 +0900)]
Change external source code download setting (#3116)

Use tar.gz file instead of zip file
Change external source code download setting to server path only
- Support smooth external code update
- Support simple mirror server moving

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years agoAdd SimpleNeg to process NEG on CPU (#2979)
윤지영/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 07:50:31 +0000 (16:50 +0900)]
Add SimpleNeg to process NEG on CPU (#2979)

* Add SimpleNeg to process NEG on CPU

This patch allows to process NEG on CPU by default

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Update if branch statement of NEG

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
* Use USE_SIMPLE_NEG environment of NEG cpu operation

It is quite many slower than gpu operation.
But for debugging, keep this codes.

* Revert an irrelevant correction

Signed-off-by: Jiyoung Yun <jy910.yun@samsung.com>
5 years ago[contrib] running/synchronizing kernels on two devices in a GPU (#2780)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 12 Oct 2018 07:42:53 +0000 (16:42 +0900)]
[contrib] running/synchronizing kernels on two devices in a GPU (#2780)

This is a experiment code to test how to synchronize kernels on two devices in a GPU.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years agoAdd bash shebang to generate_test.sh (#3102)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 07:33:17 +0000 (16:33 +0900)]
Add bash shebang to generate_test.sh (#3102)

`$BASH_SOURCE` is not available even for some other bash-compatible
shells.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoAssert moved:exec to compile time for DepthToSpace Layer (#3088)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 12 Oct 2018 07:27:27 +0000 (12:57 +0530)]
Assert moved:exec to compile time for DepthToSpace Layer (#3088)

If shape constraint has done well at compilation time, then there is no
need of assert at execution time.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoAdding Scale, Overflow policy and Rounding policy to Div op (#3086)
Siva Sai Vaddipati/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 12 Oct 2018 07:25:32 +0000 (12:55 +0530)]
Adding Scale, Overflow policy and Rounding policy to Div op (#3086)

Same as Mul op which uses the following values.

Signed-off-by: Siva Sai <siva.sai@samsung.com>
5 years agoRemove space_to_batch gtest from skiplist armv7l-linux (#3079)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 12 Oct 2018 07:23:08 +0000 (12:53 +0530)]
Remove space_to_batch gtest from skiplist armv7l-linux (#3079)

This patch will remove passing space_to_batch gtest from skiplist.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoEnable to check operand status in pureacl (#2438)
남궁석/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 06:37:59 +0000 (15:37 +0900)]
Enable to check operand status in pureacl (#2438)

This commit will enable to check if operand exists or not

Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
5 years ago[neurun] Rename IBackendConfig to IConfig (#3098)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 03:36:04 +0000 (12:36 +0900)]
[neurun] Rename IBackendConfig to IConfig (#3098)

Rename `IBackendConfig` to `IConfig` since it is in namespace
`neurun::backend` so "Backend" in the class name is redundant.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoSeperate functions of Cast.h into the declaration and the definition (#3104)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 12 Oct 2018 02:06:24 +0000 (11:06 +0900)]
Seperate functions of Cast.h into the declaration and the definition (#3104)

This commit seperates functions of Cast.h into the declaration and the definition.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years ago[neurun] Show other inputs on dot graph (#3078)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 11 Oct 2018 22:14:27 +0000 (07:14 +0900)]
[neurun] Show other inputs on dot graph (#3078)

* [neurun] Show other inputs on dot graph

This commit shows other inputs(kernel,bias..) on dot graph.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Add options to switch showing range

* merge condition

5 years agoFix max_pool op to support quant type (#3016)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 11:30:36 +0000 (17:00 +0530)]
Fix max_pool op to support quant type (#3016)

This patch will fix the max_pool_quant8_3 Gtest
And adds support for Quantization.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoMove TRANSPOSE_CONV_EX case to AddOperationEx in model.cc (#3103)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 11:29:17 +0000 (16:59 +0530)]
Move TRANSPOSE_CONV_EX case to AddOperationEx in model.cc (#3103)

This patch moves TRANSPOSE_CONV_EX case to AddOperationEx since it's an Ex operation.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[neurun] Make codegen impl less verbose (#3084)
PRAVEEN DORESWAMY NAIDU/System SW /SRI-Bangalore/Staff Engineer/삼성전자 [Thu, 11 Oct 2018 11:28:36 +0000 (16:58 +0530)]
[neurun] Make codegen impl less verbose (#3084)

Since we have the inputs and outputs as operand::Index we do not need the full
scope resolution, hence changing those to auto for codegen.

Signed-off-by: Praveen D N <praveen.dn@samsung.com>
5 years agoAdd NNAPI Delegation Changes for L2_Normalization op (#3080)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 11:27:49 +0000 (16:57 +0530)]
Add NNAPI Delegation Changes for L2_Normalization op (#3080)

This patch adds NNAPI Delegation Changes for L2_Normalization op.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoNNAPI delegation changes for Batch_To_Space_ND op (#3057)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 10:51:31 +0000 (16:21 +0530)]
NNAPI delegation changes for Batch_To_Space_ND op (#3057)

This patch gives the required NNAPI_Delegation changes for Batch_To_Space_op.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoRemove Unsupported Datatye Cases in SpaceToDepth (#3077)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 10:46:48 +0000 (16:16 +0530)]
Remove Unsupported Datatye Cases in SpaceToDepth (#3077)

NN api defines only for U8 and F32 dataype for spacetodepth op.
Hence, leaving code only for supported dataypes.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoRemove S8,U32,S32 dataype cases from DepthToSpace (#3076)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 10:46:31 +0000 (16:16 +0530)]
Remove S8,U32,S32 dataype cases from DepthToSpace (#3076)

This patch will remove S8, U32 and S32 cases since nnapi doesnot defines for them in DepthToSpace

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoAdd LRN codes in NE-ACL to supp both odd/even val (#2908)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 08:07:09 +0000 (13:37 +0530)]
Add LRN codes in NE-ACL to supp both odd/even val (#2908)

This patch contains RS7-RuntimeNTools/ComputeLibrary#47 PR
For NEON-ACL LRN support
Updated license contents for NENormalizationLayerEx

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoFix for avg_pool_quant8_4 Gtest (#3017)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 05:28:20 +0000 (10:58 +0530)]
Fix for avg_pool_quant8_4 Gtest (#3017)

This patch will fix avg_pool_quant8_4 gtest

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agonnfw: RSQRT support for Extended ACL. (#3039)
TANUJ TEKRIWAL/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 05:11:30 +0000 (10:41 +0530)]
nnfw: RSQRT support for Extended ACL. (#3039)

SQRT is supported as an Activation function in ACL, so RSQRT is also
added as in Activation Function in case of Extended ACL for nnfw.

Full framework for Activation function is added with this patch so
that if similar functionality is required for some other operator
then it can be added easily.

Signed-off-by: Tanuj Tekriwal <tanuj.tekri@samsung.com>
5 years agoIntroduce BATCH_TO_SPACE_ND in runtime. (#3058)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 05:10:23 +0000 (10:40 +0530)]
Introduce BATCH_TO_SPACE_ND in runtime. (#3058)

This patch introduces BATCH_TO_SPACE_ND in PACL runtime.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoRemove cmake for acl build (#3092)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 11 Oct 2018 04:18:23 +0000 (13:18 +0900)]
Remove cmake for acl build (#3092)

Remove cmake for acl build (not used anymore)

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years agoAdd NNAPI_Delegation changes for HashTableLookup op (#3082)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 11 Oct 2018 02:23:35 +0000 (07:53 +0530)]
Add NNAPI_Delegation changes for HashTableLookup op (#3082)

This patch adds NNAPI_Delegation changes for HashTableLookup op

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoIntroduce Custom(Abs) op in runtime (#3035)
Siva Sai Vaddipati/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 12:42:18 +0000 (18:12 +0530)]
Introduce Custom(Abs) op in runtime (#3035)

This commit introduces Custom(Abs) op in PACL runtime.
Related issue: #2828

Signed-off-by: Siva Sai <siva.sai@samsung.com>
5 years ago[PureACL] remove unused code (tanh activation code) (#3072)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 10 Oct 2018 10:54:35 +0000 (19:54 +0900)]
[PureACL] remove unused code (tanh activation code) (#3072)

Remove tanh fuseCode since we don't have tanh supported.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years agoUpdating skiplist for HashtableLookup (#2907)
Siva Sai Vaddipati/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 10:45:09 +0000 (16:15 +0530)]
Updating skiplist for HashtableLookup (#2907)

With #2655 and #2708 both the removed gtests are passing,
hence removing them from the skiplist.

Signed-off-by: Siva Sai <siva.sai@samsung.com>
5 years ago[neurun] Fix condition typo in lowering (#3071)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 10:33:01 +0000 (19:33 +0900)]
[neurun] Fix condition typo in lowering (#3071)

This commit fixes condition typo in lowering.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
5 years ago[neurun] Remove dot dump env verbose (#3073)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 10:23:30 +0000 (19:23 +0900)]
[neurun] Remove dot dump env verbose (#3073)

This commit removes dot dump env verbose which is not needed anymore.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
5 years agoFix wrong return of EnvVar::asBool (#3075)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 10 Oct 2018 10:15:10 +0000 (19:15 +0900)]
Fix wrong return of EnvVar::asBool (#3075)

Previously `asBool` always return `1`. Fixed the bug.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
5 years ago[neurun] Combine generateWeight and generateBias (#3066)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 08:25:51 +0000 (17:25 +0900)]
[neurun] Combine generateWeight and generateBias (#3066)

Combine `generateWeight` and `generateBias` into one `generate` method
in `IInitializerGenerator`.

The type `Initializer` is now a list of a pair of Index and Initializer
function. To keep the legacy code as much as possible, the two functions
are still there as private and non-virtual.

Resolve #3042

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoAdd CL Kernels for SquaredDifference op. (#3037)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 08:04:59 +0000 (13:34 +0530)]
Add CL Kernels for SquaredDifference op. (#3037)

This patch adds CL Kernels for SquaredDifference op.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[neurun] Move backend interfaces to interface dir (#3041)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 05:28:55 +0000 (14:28 +0900)]
[neurun] Move backend interfaces to interface dir (#3041)

Move backend interfaces to interface dir.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoSetting quantization info to fix bug (#3020)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 05:04:36 +0000 (10:34 +0530)]
Setting quantization info to fix bug (#3020)

This patch will set quantization info for the quant8 supported ops.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoRemove passed Generated Tests from skiplist armv7l-linux (#3049)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 04:49:12 +0000 (10:19 +0530)]
Remove passed Generated Tests from skiplist armv7l-linux (#3049)

This patch removes the passed Generated Tests
```
tensorflowmax_ex*
space_to_depth*
```
from the skiplist ```runtime_run_android_nn_test.skip.armv7l-linux```.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoAdd Generated Test for Squared Difference op (#3019)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 04:37:28 +0000 (10:07 +0530)]
Add Generated Test for Squared Difference op (#3019)

Add squared_difference_ex_1D_float for testing 1d tensor.
Add squared_difference_ex_2D_float for testing 2d tensor.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoIncluded file is not required (#2910)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 03:05:48 +0000 (08:35 +0530)]
Included file is not required (#2910)

This patch will remove #include CLArithmeticSubtraction which is not getting used at all.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoChanges in Generated files of gather gTests (#3045)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 03:05:12 +0000 (08:35 +0530)]
Changes in Generated files of gather gTests (#3045)

The files
```
gather_1D_quant8.example.cpp
gather_2D_quant8.example.cpp
gather_1D_quant8.model.cpp
gather_2D_quant8.model.cpp
```
have changes that is not reflected in the nnfw repo.
This patch points out those necessary changes.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[neurun] Only one item for codegen operand Context (#3043)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 02:53:43 +0000 (11:53 +0900)]
[neurun] Only one item for codegen operand Context (#3043)

* [neurun] Only one item for codegen operand Context

Now that Permutation insertion has changed that there must be only one
`backend::operand::Object` for each operand. So this commit removes
vector container which made it one-to-many relationship between graph
operand and backend operand.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Add assertion

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoAdd NNAPI Delegation and Custom Op for Equal (#2984)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 02:33:26 +0000 (08:03 +0530)]
Add NNAPI Delegation and Custom Op for Equal (#2984)

This patch adds NNAPI Delegation and Custom op definition for Equal.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoRemove depth_to_space gtest from skiplist armv7l-linux (#3052)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 02:10:49 +0000 (07:40 +0530)]
Remove depth_to_space gtest from skiplist armv7l-linux (#3052)

This patch will remove passing depth_to_space gtest from skiplist.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoImplementation of Space_to_BatchND op in PACL (#2854)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 02:10:00 +0000 (07:40 +0530)]
Implementation of Space_to_BatchND op in PACL (#2854)

This patch will acc CPU implemenation of SPACETOBATCHND for rank=4 tensors.

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years ago[neurun] Simplify graph dot dumper implementation (#3034)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 02:07:34 +0000 (11:07 +0900)]
[neurun] Simplify graph dot dumper implementation (#3034)

* [neurun] Simplify graph dot dumper implementation

This commit unifies DotBuilder update methods into one, and remove
redundant shape info in the parameter list.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Simplify addNode

5 years agoAdd Apache License (#3005)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Wed, 10 Oct 2018 01:25:20 +0000 (06:55 +0530)]
Add Apache License (#3005)

This patch adds Apache License to SimpleEmbedding lookup and SquaredDiff layer

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years ago[neurun] Remove unnecessary header includes (#3053)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 10 Oct 2018 01:17:19 +0000 (10:17 +0900)]
[neurun] Remove unnecessary header includes (#3053)

Remove unnecessary header includes in
`frontend/wrapper/compilation.cc`.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years ago[neurun] Update Plan and TensorBuilder role (#3011)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 9 Oct 2018 23:56:55 +0000 (08:56 +0900)]
[neurun] Update Plan and TensorBuilder role (#3011)

Update `ITensorBuilder::prepare` not to depend on `Plan` object.

In order to do that, Introduce `wrapTensor` and `iterate` interfaces for
`ITensorBuilder`.

Resolve #2746

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoPatch bug in ArmComputeEx NEG_Ex (#3018)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Mon, 8 Oct 2018 05:25:17 +0000 (10:55 +0530)]
Patch bug in ArmComputeEx NEG_Ex (#3018)

This patch solves the bug in NEG_Ex in ArmComputeEx.
Related issue: #2920

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoFix run script (#3024)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 8 Oct 2018 02:52:41 +0000 (11:52 +0900)]
Fix run script (#3024)

This will fix run script that was incorrecly fixed

Signed-off-by: SaeHie Park <saehie.park@samsung.com>
5 years ago[neurun] Add default ctor for operand::Shape (#3025)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 8 Oct 2018 02:51:17 +0000 (11:51 +0900)]
[neurun] Add default ctor for operand::Shape (#3025)

Add default ctor for `graph::operand::Shape` to remove meaningless
initialization with rank param like `Shape{1}`.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoFix a bug of StridedSliceCPU (#3015)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 8 Oct 2018 01:34:57 +0000 (10:34 +0900)]
Fix a bug of StridedSliceCPU (#3015)

This commit fixes a bug of StridedSliceCPU.
- Fix StridedSliceCPU to work with `ICLTensor` even if the `ICLTensor`s have padding.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years agoFix a bug of ReduceSum (#3013)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 8 Oct 2018 01:34:42 +0000 (10:34 +0900)]
Fix a bug of ReduceSum (#3013)

This commit fixes a bug of ReduceSum.
- Fix ReduceSum to work with `ICLTensor` even if the `ICLTensor`s have padding.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years agoFix a bug of ReduceMax for cpu (#3012)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 8 Oct 2018 01:34:30 +0000 (10:34 +0900)]
Fix a bug of ReduceMax for cpu (#3012)

This commit fixes a bug of ReduceMax for cpu.
- Fix ReduceMax to work with `ICLTensor` even if the `ICLTensor`s have padding.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
5 years ago[neurun] Implement to run the dot dumper (#2912)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 5 Oct 2018 07:14:50 +0000 (16:14 +0900)]
[neurun] Implement to run the dot dumper (#2912)

* [neurun] Implement to run the dot dumper

This commit implements to run the dot dumper with env GRAPH_DOT_DUMP.

Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* add

* Extract dump method to DotDumper class

* add license to DotDumper.cc

5 years agoAdd Apache License to Local Response Normalization op (#2932)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 06:25:36 +0000 (11:55 +0530)]
Add Apache License to Local Response Normalization op (#2932)

This patch adds Apache License to Local Response Normalization op.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoIntroduce TRANSPOSE_CONV operation on runtime (#2941)
Devansh Bansal/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 06:05:26 +0000 (11:35 +0530)]
Introduce TRANSPOSE_CONV operation on runtime (#2941)

This patch introduces TRANSPOSE_CONV operation on runtime.
This operator is required by grocery recognition system model.

Signed-off-by: b.devansh <b.devansh@samsung.com>
5 years ago[neurun] Rewrite the way for addShapeConstr (#2937)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Fri, 5 Oct 2018 04:55:41 +0000 (13:55 +0900)]
[neurun] Rewrite the way for addShapeConstr (#2937)

Now that we have input/outputs as `operation::Node` interface, we do not
need to implement calls to `addShapeConstr` for each node. Instead, it
is done in general way for all nodes from `Linear::markTensors`. And for
this, there was some changes to `ITensorBuilder` interface.

Reslove #2741

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years ago[tfltool] OP selector: Support more operator (#2975)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 5 Oct 2018 04:14:51 +0000 (13:14 +0900)]
[tfltool] OP selector: Support more operator (#2975)

Support selecting operators: NEG, EXP

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
5 years agoChange input arguments of SquaredDifference operator (#2903)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 04:10:32 +0000 (09:40 +0530)]
Change input arguments of SquaredDifference operator (#2903)

While testing grocery model, below errors occured,
```
tflite_run: /home/jyoung/git/star/nnfw/runtimes/pure_arm_compute/src/internal/op/SquaredDifference.cc:33: internal::tflite::op::SquaredDifference::Param::Param(uint32_t, const uint32_t*, uint32_t, const uint32_t*): Assertion `inputCount == 3 && outputCount == 1' failed.
```
There is no need of an activation index as SquaredDifference op takes only two inputs X & Y.
This patch makes those necessary changes.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoUpdate nnapi_delegation for Space_To_BatchND (#2904)
Shubham Gupta/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 04:06:59 +0000 (09:36 +0530)]
Update nnapi_delegation for Space_To_BatchND (#2904)

This patch will add nnapi delegaion for Space_to_BatchND

Signed-off-by: shubham <shub98.gupta@samsung.com>
5 years agoAdd generated tests for NEG_EX (#2869)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 01:21:36 +0000 (06:51 +0530)]
Add generated tests for NEG_EX (#2869)

Add neg_ex_1D_float for testing 1d tensor.
Add neg_ex_2D_float for testing 2d tensor.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years agoAdd Apache License to L2Normalization Op (#2931)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Fri, 5 Oct 2018 01:20:33 +0000 (06:50 +0530)]
Add Apache License to L2Normalization Op (#2931)

This patch adds Apache License to L2Normalization Op.

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[neurun] Introduce Enum `Input` for Conv2D (#2943)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 4 Oct 2018 10:59:30 +0000 (19:59 +0900)]
[neurun] Introduce Enum `Input` for Conv2D (#2943)

Introduce Enum `Input` for FullyConnected to avoid using magic numbers
from input list.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years agoAdd Apache License to SquaredDifference Op (#2930)
Prasanna R/System SW /SRI-Bangalore/Engineer/삼성전자 [Thu, 4 Oct 2018 10:58:27 +0000 (16:28 +0530)]
Add Apache License to SquaredDifference Op (#2930)

This patch adds Apache License to Squared Difference Op

Signed-off-by: prasannar <prasanna.r@samsung.com>
5 years ago[neurun] Introduce Enum `Input` for MaxPool2D (#2945)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 4 Oct 2018 10:30:59 +0000 (19:30 +0900)]
[neurun] Introduce Enum `Input` for MaxPool2D (#2945)

Introduce Enum `Input` for MaxPool2D to avoid using magic numbers
from input list.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
5 years ago[neurun] Introduce Enum `Input` for the rest ops (#2947)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 4 Oct 2018 10:30:48 +0000 (19:30 +0900)]
[neurun] Introduce Enum `Input` for the rest ops (#2947)

Introduce Enum `Input` for Reshape and Softmax to avoid using magic
numbers from input list.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>