Refactoring of Shape::broadcast (#4132)
Remove 4 loops and creating two temp vectors and handle all checks in
one loop
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 14 Jan 2019 08:16:44 +0000 (13:46 +0530)]
Bug fix for output shape in Permute (#4201)
This patch will fix the bug in calc output shape in Permute
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 14 Jan 2019 01:23:26 +0000 (06:53 +0530)]
Add more validations in Permute CL host file (#4188)
This patch will add validations to check input
and permutation vector.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Sun, 13 Jan 2019 21:50:32 +0000 (03:20 +0530)]
Correction in variable names in CL Arg op files (#4125)
Since somewhere in code variable name is argminmax_axis and
somewhere it is just axis along with comments having name as
axis.So, this patch will corrects the comments and replaces
all argminmax_axis to axis to have a uniformity in variable
name for axis.
Signed-off-by: shubham <shub98.gupta@samsung.com>
김수진/On-Device Lab(SR)/Engineer/삼성전자 [Sun, 13 Jan 2019 21:48:36 +0000 (06:48 +0900)]
[neurun] Support broadcast for ADD op (#4203)
Related : #4173
This commit supports broadcast for ADD op and enable the gtest.
Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
Synch tools/nnapi_quickcheck with current changes (#4197)
Since we don't build tools/nnapi_quickcheck, it's far behind from
current state. So, this change updates all files of this dir to be able
to build and run them
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
[neurun] Fix CLTensor::setBuffer to work with new acl (#4199)
Related issue: 4185
In new versions of acl that we use now, its functions used in this
method were either updated or removed. This change updates
CLTensor::setBuffer to be able to address these new changes
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Fri, 11 Jan 2019 01:00:29 +0000 (06:30 +0530)]
Remove unused func from CLArgMinMaxKernel file (#4200)
This patch will remove unused func from CLArgMinMaxKernel.h
Signed-off-by: shubham <shub98.gupta@samsung.com>
김수진/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Jan 2019 00:50:18 +0000 (09:50 +0900)]
[neurun] Enable basic ADD op (#4186)
* [neurun] Enable basic ADD op
Related : #4173
This commit migrate basic `ADD` op that is already exist in `PACL` to `neurun`.
- Neon cannot be supported now.
- TensorBuilder should has not only `arm_compute::CLTensor` but also `arm_compute::Tensor`.
- Remove skip test `GeneratedTests.add`
Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* Remove neon codes
* Remove unused codes
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Jan 2019 00:45:01 +0000 (09:45 +0900)]
[neurun] Extract Operand LowerInfo from Node (#4184)
* [neurun] Extract Operand LowerInfo from Node
As `model::operand::Node` is pure node info so LowerInfo should not be
Node class. `graph::Graph` holds them instead.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Do not pass Graph object for DotOperandInfo
Pass only `LowerInfo` instead
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
김수진/On-Device Lab(SR)/Engineer/삼성전자 [Fri, 11 Jan 2019 00:33:27 +0000 (09:33 +0900)]
[PACL] Extract internal kernels to libs (#4144)
* [PACL] draft: Extract internal kernels to libs
Related : #4060
- Introduce `libs/kernels` and `libs/kernels/simple`
- Move `ARMComputeEx` to `libs/kernels`
- Move internal kernels of `PACL` to `libs/kernels/simple`
- Remove `FeatureLoggingLayer`
- Some utils are copied from `PACL` to `libs/kernels/simple`
Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
* remove *.clembed
* Revert kernels location
* Fix build failure
* Change directory to misc
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Jan 2019 08:14:36 +0000 (17:14 +0900)]
Define more skiplist and install (#4195)
Define more skiplist: neurun x86, neurun armv7l cpu
Install all skiplists for target platform
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Jan 2019 04:38:21 +0000 (13:38 +0900)]
Remove warning from acl on gbs build (#4194)
Remove ignored-attributes warning from acl ex header on gbs build
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Jan 2019 03:59:04 +0000 (12:59 +0900)]
Use more strict compile option for misc and profiling (#4193)
Use more strict compile option for misc and profiling
Remove use of size_t as possible
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Thu, 10 Jan 2019 01:10:38 +0000 (06:40 +0530)]
Correcting name to asPadStrideInfo from asPadStringInfo (#4190)
Since by asPadStringInfo, one cant think if it is having
stride info or not. Hence, this patch wil rename this func
name to asPadStrideInfo.
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 10 Jan 2019 01:07:23 +0000 (10:07 +0900)]
Strict build option for acl (#4163)
* Strict build option for acl
- Add build option for more strict
- Except warning unused-parameter and unused-function for validation check
- Add assert, log, debug build option for debug configure
- Fix bug
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Fix validatoin check & broadcast bug
* Fix update_window_and_padding call using or operation
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 10 Jan 2019 00:28:07 +0000 (09:28 +0900)]
[neurun] Remove `Index::asInt` (#4189)
Remove `asInt` method from `Index` since it should return its underlying
type. Calls to `asInt()` has been replaced with `value()`.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 9 Jan 2019 09:32:29 +0000 (18:32 +0900)]
Enable neurun x86 build (#4187)
Enable neurun x86 build - use cpu kernel only
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 9 Jan 2019 07:54:45 +0000 (16:54 +0900)]
[neurun] Do not use raw int in operand::Context (#4182)
This commit revises the type of key for `compiler::operand::Context` to
be `Index` class rather than raw int. Since Index does not support
`operator<` the conainer became `unordered_map`.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 9 Jan 2019 07:28:50 +0000 (16:28 +0900)]
Change macro `DEBUG` to `ARM_COMPUTE_DEBUG_ENABLED` (#4183)
This commit changes `DEBUG` option to `ARM_COMPUTE_DEBUG_ENABLED` in ARMComputeEx.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Wed, 9 Jan 2019 04:32:05 +0000 (13:32 +0900)]
[neurun] Move Backend info to PermuteNode (#4180)
In order for StageGenerator to not have dependency on backends.
This commit moves the backend info for input/output from StageGenerator
to creation of PermuteNode.
(which is created only in PermutationInsertionPass)
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 9 Jan 2019 04:15:33 +0000 (13:15 +0900)]
Use -fPIC build flag for tflite only (#4175)
* Use -fPIC build flag for tflite only
Instead of global setting, use -fPIC build flag for tenroflow lite only
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Specify to build tensorflow lite library static
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 9 Jan 2019 00:51:25 +0000 (06:21 +0530)]
Correction in validation func in CL Host files (#4139)
Since ARM_COMPUTE_RETURN_ERROR_ON_MSG() is an "ERROR_ON" macro,
which means an error will be thrown if the expression evaluates to "true"
Hence, checks need to be inverted.
This patch will correct the checks in th following files:
1. CLBatchToSpaceNDKernel.cpp
2. CLDepthToSpaceKernel.cpp
3. CLPadLayerKernel.cpp
4. CLReduceOperationKernel.cpp
5. CLSpaceToDepthKernel.cpp
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 9 Jan 2019 00:48:51 +0000 (09:48 +0900)]
Remove warning from external acl and tensorflow (#4181)
Remove warning from external acl and tensorflow by using SYSTEM keyword
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 8 Jan 2019 10:12:50 +0000 (19:12 +0900)]
Use acl pre-built binary (#4179)
Remove workaround to avoid build fail by acl version update
Use pre-built acl v18.11 binary
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
장지섭/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Jan 2019 07:10:18 +0000 (16:10 +0900)]
Update version of ARMComputeEx to 18.11 (#3877)
* Update version of ARMComputeEx to 18.11
This commit updates the version of ARMComputeEx to 18.11.
- Change the name of the kernel of StridedSlice and PadLayer to StridedSliceEx and PadLayerEx to prevent duplicated name.
- Remove the code related to fixed point(QS8, QS16).
- Update CLKernelLibraryEx to 18.11.
- Apply using the same context for CLKernelLibraryEx as CLScheduler.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
* Change config files to build with acl v18.11
This commit changes config files to build with acl v18.11.
- Change download url of acl from v18.05 to v18.11
- Change acl to always build
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
* Enable the build of neurun with v18.11
This commit enables the build of neurun with v18.11.
For detail : https://github.sec.samsung.net/STAR/nnfw/pull/3877#issuecomment-227170
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
* acl benchmark patch
* Fix missing cl kernel function
This commit adds missing cl kernel function to map list.
Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 8 Jan 2019 03:42:05 +0000 (12:42 +0900)]
Remove acl link for neurun ineternal (#4174)
Remove acl link for neurun ineternal
Becasue neurun internal is independent with acl, we can remove acl link
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 8 Jan 2019 03:41:40 +0000 (12:41 +0900)]
Environment variable to disable compile (#4172)
Introduce environment variable to disable compile
Return with bad state in exection phase when compile state is NOT_COMPILED
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Jan 2019 01:08:19 +0000 (10:08 +0900)]
[neurun] Move Permute Type struct to PermuteNode (#4166)
In order for StageGenerator to not have dependency on backends.
This commit moves Permute Type decision from StageGenerator to creation
of PermuteNode(which is created only in PermutationInsertionPass).
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 8 Jan 2019 01:03:15 +0000 (10:03 +0900)]
[neurun] Remove StageGenerator log prints (#4167)
Remove StageGenerator log prints since it is not useful anymore.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 7 Jan 2019 08:42:11 +0000 (17:42 +0900)]
Revert acl cmake build (#4164)
Revert acl cmake build to update acl v18.11
This reverts commit
5e4e7e27d7b4a20ef300bb4577c4e14369db7e06.
This reverts commit
7d484e5c3acd4b3ad8039226a25f78d26748be0a.
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
이상규/On-Device Lab(SR)/Principal Engineer/삼성전자 [Mon, 7 Jan 2019 02:50:12 +0000 (11:50 +0900)]
Add pause time between each tests in benchmark-op (#4158)
To reduce the impact of high temperature due to prior running,
this patch will add pause time between each run.
Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 7 Jan 2019 02:16:30 +0000 (07:46 +0530)]
Correcting assertion in transposeConv CPU implementation (#4153)
This patch will correct the position of bracket which has
been wrongly placed.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Fri, 4 Jan 2019 03:09:36 +0000 (08:39 +0530)]
Remove unnecessary headers from CLReduce and CLArg ops (#4124)
This patch will remove unnecessary headers from CLReduce
and CLArg op files. And Replace CLArray to CLScheduler
since only CLScheduler header file is enough.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Fri, 4 Jan 2019 01:23:29 +0000 (06:53 +0530)]
Remove unused variables in CLArgMinMaxKernel file (#4127)
This patch will remove unused variables from
CLArgMinMaxKernel.cpp
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Fri, 4 Jan 2019 01:21:42 +0000 (10:21 +0900)]
Block gtest install (#4146)
- Update add_extdirectory macro for EXCLUDE_FROM_ALL setting
- Disable gtest install using add_subdirectory with EXCLUDE_FROM_ALL
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Removing ambiguity from pack gtests (#4074)
This patch rectifies the name of one of the inputs.
Signed-off-by: b.devansh <b.devansh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 3 Jan 2019 10:16:13 +0000 (19:16 +0900)]
Update docker builder script (#4141)
- Fix to handle multiple docker image build argument
- Set default image name
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 3 Jan 2019 08:38:33 +0000 (17:38 +0900)]
Update cmake for revised directory structure (#4143)
Update cmake for android and gbs build for revised directory structure
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Adding quant8 gtest for split (#4046)
This patch adds quant8 gtest for split.
Signed-off-by: b.devansh <b.devansh@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Thu, 3 Jan 2019 06:06:36 +0000 (15:06 +0900)]
[neurun] Extract Operation LowerInfo from Node (#4136)
As `model::operation::Node` is pure node info so LowerInfo should not
be Node class. `graph::Graph` holds them instead.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 2 Jan 2019 10:24:18 +0000 (19:24 +0900)]
Remove workaround for test script path change (#4134)
Remove workaround for test script path change because CI scripts are updated
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 2 Jan 2019 07:10:46 +0000 (16:10 +0900)]
Update validation test skiplist (#4133)
Remove model vaildation tests in neurun skiplist
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Clean up dependencies of backends (#4112)
After resolving dependencies and converting backends to plugins, these
dependencies aren't actual anymore
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
Remove nnapi library (#4105)
Related issue: 4103
Remove nnapi library since just one of its functions is used in
backends: better to move it into neurun/utils
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 2 Jan 2019 02:26:50 +0000 (07:56 +0530)]
Changes in cl kernel files related to #if defined conditions (#4106)
1. Add missing #if defined conditions
2. Remove the Vec_size from #if defined check since kernel
should execute even if it is not passed as an preprocessor
argument with an assumption VEC_SIZE=1 i.e., processing
on one element at a time.
Files modified are:
1. embedding_lookup.cl
2. exp.cl
3. hashtable_lookup.cl
4. neg_tensor.cl
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 2 Jan 2019 02:23:16 +0000 (07:53 +0530)]
Changes in Cast op files (#4104)
1. Add missing comments in Cast kernel
2. Add #if defined to have error handling
3. Renaming of OFFSET_IN/SCALE_IN to offset and scale
to remove ambiguity if it is for input or output
due to the variable name
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 2 Jan 2019 02:21:11 +0000 (07:51 +0530)]
Add missing condition in Batch_to_space kernel for #if defined (#4101)
This patch will add one more condition to check if defined(Batch_OUT).
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 2 Jan 2019 02:20:28 +0000 (07:50 +0530)]
Corrections in comments and adding #if defined for PRELU kernel (#4100)
This patch will corrects the comments and add #if defined for PRELU kernel
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 2 Jan 2019 02:17:54 +0000 (11:17 +0900)]
Revise compile state (#4086)
Move lower, linearize state setting to plan
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 2 Jan 2019 02:17:25 +0000 (07:47 +0530)]
Add QASYMM8 support for PReLU op (#4079)
This patch adds QASYMM8 support for PReLU op.
Issue: #3884
Signed-off-by: prasannar <prasanna.r@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 24 Dec 2018 05:04:36 +0000 (10:34 +0530)]
Arrange GTests in linux skip lists in order (#4092)
Gtest in skiplist were in order before #3878
This patch arranges the same for better organizing and tracking
Signed-off-by: prasannar <prasanna.r@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 24 Dec 2018 05:03:32 +0000 (14:03 +0900)]
Fix build for svace and coverity (#4121)
Fix build for svace and coverity: fix for renamed generated test
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
박종현/On-Device Lab(SR)/Staff Engineer/삼성전자 [Mon, 24 Dec 2018 05:03:04 +0000 (14:03 +0900)]
Fix ARMCompute build script bug (#4097)
The current version is unable to build ARMCompute from the downloaded
source.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
[neurun] Load backend dynamically as plugin (#3863)
Load backends dynamically as plugins during runtime
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Fri, 21 Dec 2018 01:07:18 +0000 (06:37 +0530)]
Removing unnecessary preprocessor argument in StridedSlice op (#4102)
Since Element_size is not getting used in the kernel, hence
this patch will remove this unnecessary preprocessor argument
for Element_size
Signed-off-by: shubham <shub98.gupta@samsung.com>
[neurun]Merge 2 for-loops of PlanBuilder::finalize (#4118)
Since these two loops are independent and loop over the same
tensor_builders, better to merge them
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
[neurun] CamelCase DotDumper::dump_if_needed (#4117)
Convert style of DotDumper::dump_if_needed to camel case
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 20 Dec 2018 22:19:33 +0000 (07:19 +0900)]
Calculate build depth before script backup (#4107)
Calculate build depth and make depth count text file
Prepare CI update and removing backup
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Thu, 20 Dec 2018 01:45:14 +0000 (10:45 +0900)]
Cmake for gbs build fix (#4115)
Fix gbs build failure: changed nnapi gtest directory name
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Thu, 20 Dec 2018 00:55:19 +0000 (06:25 +0530)]
Add #ifndef block for VEC_SIZE in comparison op cl kernel (#4109)
This patch adds #ifndef block for VEC_SIZE in comparison op cl kernel
Signed-off-by: prasannar <prasanna.r@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Thu, 20 Dec 2018 00:54:55 +0000 (06:24 +0530)]
Changes req in Binary Logical op files (#4082)
1. Add op_code related comment
2. Initialize op_code variable to have error handling
3. Indentation and correction in already present commnets
changing #if defined conditions
Signed-off-by: shubham <shub98.gupta@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Thu, 20 Dec 2018 00:50:50 +0000 (06:20 +0530)]
Add #endif to #ifndef block in comparison_op_quantized cl (#4110)
This patch adds #endif to #ifndef block in comparison_op_quantized cl
Else it would be a fatal error stating unterminated #ifndef
Signed-off-by: prasannar <prasanna.r@samsung.com>
Devansh Bansal/SNAP /SRI-Bangalore/Engineer/삼성전자 [Thu, 20 Dec 2018 00:49:14 +0000 (06:19 +0530)]
Adding runtime asserts in TransposeConv. (#4099)
This patch adds shape related asserts for transpose_conv.
Signed-off-by: b.devansh <b.devansh@samsung.com>
[neurun] Refactory call of DotDumper in compile() (#4113)
No need to complicate Compiler::compile with codes, that don't relate to
compilation. So, this change moves env checking DotDumber related
branching into DotDumper
Signed-off-by: Poshshoev Dilshodzhon <d.poshshoev@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Wed, 19 Dec 2018 00:58:29 +0000 (09:58 +0900)]
Rename generated test (#4096)
* Rename generated test
Rename generated test binary: nnapi_gtest
Rename directory including generated test: tests/nnapi
Revise directory structure
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Fix typo
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 19 Dec 2018 00:42:50 +0000 (06:12 +0530)]
Correction of Datatypes supported by SquareDiff op (#4070)
This patch will remove S16/S32 validations and comment
since it is not supported as of now by the current
implementation.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 19 Dec 2018 00:41:34 +0000 (06:11 +0530)]
Changes in Reduce op files (#4081)
1. Add missing op_code comment
2. Initialise op_code variable to have error handling
3. Indenting kernel code
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 19 Dec 2018 00:41:03 +0000 (06:11 +0530)]
Changes in Comparison op files (#4080)
1. Add missing comments for op_code
2. Add #if definition to verify the preprocessors before executing
3. initialise op_code = 0 and changing the op_code values assigned
to operation to have runtime error handling
4. Indentation of the kernel code.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Wed, 19 Dec 2018 00:40:37 +0000 (06:10 +0530)]
Unified ArgMin and ArgMax kernel (#4083)
This patch will create a single kernel which can
handle both ArgMax and ArgMin op, hence reducing the
code redundancy.
Signed-off-by: shubham <shub98.gupta@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 18 Dec 2018 10:44:43 +0000 (19:44 +0900)]
[tflite_run] Do not support image inputs (#4098)
The image file as input is not our way of testing any longer. Since this
only support inception v3 IFM, it would be nicer not to have this
option.
Resolve #1206
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 10:14:08 +0000 (15:44 +0530)]
Redefine LogicalOperation as BinaryLogicalOperation in TypesEx and remove NOT (#4089)
This patch renames LogicalOperations as BinaryLogicalOperations in TypesEx.h
Removes NOT from enum.
Reason:
This enum was introduced with an aim of providing unified kernels for LogicalOps
But unified kernel is provided only for BinaryLogicalOps
and hence LogicalNot enum value is never used.
Hence it would be better to rename and remove NOT flag from it,
Signed-off-by: prasannar <prasanna.r@samsung.com>
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 18 Dec 2018 10:12:13 +0000 (19:12 +0900)]
[neurun] Move operation files from graph to model (#4095)
* [neurun] Move operation files from graph to model
Move pure model related classes from `graph/operaation` to
`model/operation`.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Update ifdef guards
* Fix ifdef guards
이한종/On-Device Lab(SR)/Engineer/삼성전자 [Tue, 18 Dec 2018 10:08:33 +0000 (19:08 +0900)]
[neurun] Move logging.h to `util` dir (#4087)
* [neurun] Move logging.h to `util` dir
Move logging.h to `util` dir and update namespace accordingly.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
* Fix namespace
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 07:07:43 +0000 (12:37 +0530)]
Add calls to CPU version of SimpleArgMinMax op (#4029)
This patch will add call to CPU version of ArgMinMax op
and changing SimpleArgMinMax files to use ArgOperation enum
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 [Tue, 18 Dec 2018 06:48:14 +0000 (15:48 +0900)]
Move generated test into tests directory (#4090)
Move generated test in runtimes/tests to tests/runtime directory
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 04:47:17 +0000 (10:17 +0530)]
Making comparison_op_quantized.cl generic with VEC_SIZE (#4042)
This patch makes comparison op quantized.cl more generic based on VEC_SIZE
replacement from 16 -> vec_size
This is done for better scaling of this kernel.
Signed-off-by: prasannar <prasanna.r@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 02:45:04 +0000 (08:15 +0530)]
Remove CL implementation of Equal & Not Equal (#4069)
Since CLComparisonOp which is an unified CL for comparison op is merged
the redundant files can be removed.
Related issue: #3851
Signed-off-by: prasannar <prasanna.r@samsung.com>
박종현/On-Device Lab./Staff Engineer/삼성전자 [Tue, 18 Dec 2018 02:23:06 +0000 (11:23 +0900)]
Introduce Nonius CMake Package (#4068)
This commit integrates nonius benchmark framework as a nnfw
external dependency package.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
오형석/On-Device Lab./Staff Engineer/삼성전자 [Tue, 18 Dec 2018 02:20:22 +0000 (11:20 +0900)]
Fix format and license in include (#4064)
Fix format of NeuralNetworksEx.h and NeuralNetworksExShim.h
Fix format of NeuralNetworksLoadHelpers.h: minimize diff with upstream code
Fix license of NeuralNetworksLoadHelpers.h: This file is comes from tensorflow, not android
Add comment in NeuralNetworksShim.h and NeuralNetworksLoadHelpers.h: related upstream code
Modify format checker to ignore NeuralNetworks.h
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
오형석/On-Device Lab./Staff Engineer/삼성전자 [Tue, 18 Dec 2018 02:11:50 +0000 (11:11 +0900)]
Fix coverage report script (#4085)
Fix test tool directory
Remove remove-matching not existed: runtime/nn
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:42:11 +0000 (07:12 +0530)]
Add default block in CLReduceOperation kernel (#4078)
This patch adds default block in CLReduceOperation kernel
Signed-off-by: prasannar <prasanna.r@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:40:48 +0000 (07:10 +0530)]
Add default block to CLComparisonOp CL Kernel (#4077)
This patch adds default block to CLComparisonOp CL Kernel
Signed-off-by: prasannar <prasanna.r@samsung.com>
이한종/On-Device Lab./Engineer/삼성전자 [Tue, 18 Dec 2018 01:40:20 +0000 (10:40 +0900)]
[neurun] Move operand files from graph to model (#4071)
Move pure model related classes from `graph/operand` to `model/operand`.
Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:14:48 +0000 (06:44 +0530)]
Remove Passing logicaOr gtest from skiplist (#3960)
Remove Passing logicaOr gtest from skiplist
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:08:00 +0000 (06:38 +0530)]
Add gtests to test Pad op for 2D/3D tensors (#3946)
This patch will add Pad gtest to test for 2D/3D tensors
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:07:28 +0000 (06:37 +0530)]
Add gtests for ArgMax op to test int32 dataype (#4034)
Since the kernel states it support int32 datatype
So, these test cases will verfiy the same.
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Tue, 18 Dec 2018 01:06:55 +0000 (06:36 +0530)]
Add gtests for transposeConv op to test Int32 dataype (#4037)
This patch will add gtest to test TransposeConv op
for int datatype
Signed-off-by: shubham <shub98.gupta@samsung.com>
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 10:33:34 +0000 (19:33 +0900)]
Configure ARMCompute correctly (#4066)
This commit updates Makefile and ARMComputeConfig.cmake to correctly
configure ARMCompute library targets even for CI build.
Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 17 Dec 2018 10:18:48 +0000 (15:48 +0530)]
Correction in datatypes supported in NeuralNetworksEx (#4017)
This patch will add INT32 supported comment in
1. arg_max
2. pack and unpack
3. abs
And remove INT32 supported comment from SquaredDifference
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 17 Dec 2018 09:48:49 +0000 (15:18 +0530)]
Optimising ArgMinMax kernel: remove extra global memeroy access (#4020)
This patch will optimise ArgMinMax kernel by remove the extra global memory
access which can done by local variable.
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 09:47:56 +0000 (18:47 +0900)]
Define compile state (#4055)
* Define compile state
- Define 4 compile state getter/setter in plan
- Set compiled state when compile finished
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Rename getter & setter
* Use class enum instead of enum
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 09:47:44 +0000 (18:47 +0900)]
Move test-related codes in tools into tests (#4058)
Move test-related codes in tools into tests
- nnapi_test
- tflite_run
- tflite_benchmark
- tflite_benchmark_model
Update cmake, root makefile, and spec file
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 17 Dec 2018 09:34:06 +0000 (15:04 +0530)]
Add gtests to test pack and unpack op for int daatypes (#4038)
This patch will add gtests to test pack and unpack
op for int32 datatypes
Signed-off-by: shubham <shub98.gupta@samsung.com>
Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 17 Dec 2018 09:33:44 +0000 (15:03 +0530)]
Rearranging non-class func to seperate from CLass func (#4051)
It will be better to put non-class func in namespace
to have clear seperation of class and non-class func
This patch will rearrange the val_arg func so as to
have that in a namespace.
Signed-off-by: shubham <shub98.gupta@samsung.com>
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 08:06:22 +0000 (17:06 +0900)]
Collect implementation test code into contrib/labs (#4025)
Collect contribs build by BUILD_LABS into contrib/labs
Move tools/tflite_examples into contrib/labs
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 17 Dec 2018 06:27:08 +0000 (15:27 +0900)]
[neurun] Remove backend depenency in Permute (#4011)
This commit removes backend dependency in `Permute` such as `Window`, `Iterator` and `ICLTensor`...
Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 04:21:57 +0000 (13:21 +0900)]
Remove environment checking in profiling library (#4053)
* Remove environment checking in profiling library
Remove Sync class
Define boolean sync field and setter in Context
Remove environment checking in profiling library
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
* Rename env: PROFILING_OP_SYNC
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Mon, 17 Dec 2018 04:20:41 +0000 (13:20 +0900)]
Move build flag for benchmark model tool checking (#4056)
Move build flag BUILD_TFLITE_BENCHMARK_MODEL checking into tool's cmake
Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
Prasanna R/SNAP /SRI-Bangalore/Engineer/삼성전자 [Mon, 17 Dec 2018 01:56:25 +0000 (07:26 +0530)]
Remove PReLU Layer from internal/layers (#4043)
This patch removes PReLU layer from internal/layers
Since cl kernel for the same has been implemented and merged.
Signed-off-by: prasannar <prasanna.r@samsung.com>