platform/core/ml/nnfw.git
6 years agoFix build error std::function on pure acl runtime (#1505)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 1 Jun 2018 06:32:52 +0000 (15:32 +0900)]
Fix build error std::function on pure acl runtime (#1505)

I include <functional> explicitly for std::function.

Related Issue: #1503

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoRemove ARMCL_DIR setting from arm-ubunut pureacl build script (#1504)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 06:31:28 +0000 (15:31 +0900)]
Remove ARMCL_DIR setting from arm-ubunut pureacl build script (#1504)

This commit removes ARMCL_DIR from docker_build_cross_arm_pureacl.sh

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoDefine Max namespace in pure acl runtime (#1506)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 06:29:18 +0000 (15:29 +0900)]
Define Max namespace in pure acl runtime (#1506)

Define Max namespace, node, param in pure acl runtime

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[nnapi_test] Support int32 input generation (#1498)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 06:14:34 +0000 (15:14 +0900)]
[nnapi_test] Support int32 input generation (#1498)

With this commit, nnapi_test is able to generate int32 input tensors.

Note that random generation is not supported, yet.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoShow used NN runtime filepath (#1496)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 06:13:39 +0000 (15:13 +0900)]
Show used NN runtime filepath (#1496)

This commit shows the filepath of NN runtime being used.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoSet addOperationEx return value in pure acl runtime (#1507)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 06:05:46 +0000 (15:05 +0900)]
Set addOperationEx return value in pure acl runtime (#1507)

Set addOperationEx return value in pure acl runtime
To avoid conflict, commit small change.

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[NNAPI unittest] Add 'add_2' case (#1464)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 05:51:04 +0000 (14:51 +0900)]
[NNAPI unittest] Add 'add_2' case (#1464)

* [NNAPI unittest] Add 'add_2' case

This commit adds 'add_2' unittest which takes RHS as an input, but LHS
is a constant.

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

6 years agoType-based tensor output compare (#1499)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 05:48:28 +0000 (14:48 +0900)]
Type-based tensor output compare (#1499)

This commit introduces type-based tensor output compare infrastructure
in TfLiteInterpMatchApp.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Gather OP] add nnapi delegate for Gather operation (#1465)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 1 Jun 2018 04:06:13 +0000 (13:06 +0900)]
[Gather OP] add nnapi delegate for Gather operation (#1465)

This commit adds nnapi delegate for Gather operation.
-axis parameter is added

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years agoAdd script to build pure_arm_compute runtime (#1484)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 03:27:11 +0000 (12:27 +0900)]
Add script to build pure_arm_compute runtime (#1484)

* Add script to build pure_arm_compute runtime

1. Add script to cross-build pure_arm_compute runtime for both arm-ubunut and tizen

To run test, please set LD_LIBRARY_PATH=Product/lib/pureacl instead of
Product/lib.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoFlexible input generation switch for RandomTestRunner (#1495)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 03:17:53 +0000 (12:17 +0900)]
Flexible input generation switch for RandomTestRunner (#1495)

This commit revises RandomTestRunner to make it easy to extend supported
input type.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoMore tests for STRIDED_SLICE_EX (#1488)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 02:56:01 +0000 (11:56 +0900)]
More tests for STRIDED_SLICE_EX (#1488)

Add more test for STRIDE_SLICE_EX
It's similar with STRIDE_SLICE (Fix argument)

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[support.tflite] Generic TensorView (#1494)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 02:35:13 +0000 (11:35 +0900)]
[support.tflite] Generic TensorView (#1494)

This commit revises TensorView class as a generic class.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoDetect NNAPI load failure (#1489)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 02:02:57 +0000 (11:02 +0900)]
Detect NNAPI load failure (#1489)

This commit revises NeuralNetworksShim.h and NeuralNetworksExShim.h to
abort execution when NNAPI functions are unavailable, but used.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Topkv2 OP] add nnapi delegate for Topk_v2 operation (#1463)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Fri, 1 Jun 2018 01:18:30 +0000 (10:18 +0900)]
[Topkv2 OP] add nnapi delegate for Topk_v2 operation (#1463)

* [Topkv2 OP] add nnapi delegate for Topk_v2 operation

This commit add nnapi delegate for Topk_v2 operation.
-change output order (output_value is first and output_index is second)

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Topkv2 OP] add nnapi delegate for Topk_v2 operation

This commit add nnapi delegate for Topk_v2 operation.
-change output order (output_value is first and output_index is second)

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[tflite_benchmark] Show benchmark configurations (#1491)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 01:05:30 +0000 (10:05 +0900)]
[tflite_benchmark] Show benchmark configurations (#1491)

This commit revises tflite_benchmark to show benchmark configurations
before running benchmark.

Currently, the following configurations will be shown:
 - The number of threads to be used
 - Whether NNAPI is used, or not

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[PureCL] Support DIV operation (#1482)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 00:38:50 +0000 (09:38 +0900)]
[PureCL] Support DIV operation (#1482)

* [PureCL] Support DIV operation

For #1338 and #1366

This patch adds DIV operation to the pure ACL runtime.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoUpdate Android build scripts (#1487)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Fri, 1 Jun 2018 00:18:18 +0000 (09:18 +0900)]
Update Android build scripts (#1487)

This commit revises 'build_android_rootfs.sh' to create ndk directory
for tensorflow lite buiild, and updates tensorflow lite CMakeLists.txt
to access this NDK during build.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[tflite_run] Generate random input (#1477)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 31 May 2018 11:49:14 +0000 (20:49 +0900)]
[tflite_run] Generate random input (#1477)

When running `tflite_run` without input argument it runs with values
with 0. This commit will fill the input tensors with random values
instead, same way with `nnapi_test` does.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoSubmodule update on internal_acl_build (#1483)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 11:48:27 +0000 (20:48 +0900)]
Submodule update on internal_acl_build (#1483)

`make acl` does not work on fresh repo. This commit makes submodule update on
target internal_acl_build.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years ago[Gather Op] add axis to runtime_run_android_nn_test (#1479)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 11:46:10 +0000 (20:46 +0900)]
[Gather Op] add axis to runtime_run_android_nn_test (#1479)

This commit adds axis to runtime_run_android_nn_test.
-axis parameter is added newly.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[Pure CL] Introduce Const Tensor View (#1474)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 08:43:27 +0000 (17:43 +0900)]
[Pure CL] Introduce Const Tensor View (#1474)

This commit introduces ConstView class to make it easy to write generic
const operand updates.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoUpdate comment for STRIDED_SLICE in NeuralNetworks.h (#1460)
최형규/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 07:56:29 +0000 (16:56 +0900)]
Update comment for STRIDED_SLICE in NeuralNetworks.h (#1460)

- Update comment for STRIDED_SLICE in NeuralNetworks.h,
  because operands of STRIDED_SLICE has been implemented by
  following parameters defined in the latest Android NN API.

Signed-off-by: Hyung-Kyu Choi <hk0110.choi@samsung.com>
6 years ago[Pure CL] Fix typo (USE_CL_ARITHMETIC_ADDTION) (#1480)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 07:56:03 +0000 (16:56 +0900)]
[Pure CL] Fix typo (USE_CL_ARITHMETIC_ADDTION) (#1480)

This commit fixes type in environment variable name:
(USE_CL_ARITHMETIC_ADDTION -> USE_CL_ARITHMETIC_ADDITION)

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoset ROOTFS_DIR on docker_build_cross_arm_ubuntu.sh (#1467)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 07:49:03 +0000 (16:49 +0900)]
set ROOTFS_DIR on docker_build_cross_arm_ubuntu.sh (#1467)

To run docker_build_cross_arm_ubuntu, ROOTFS_DIR should be passed.
It takes 6min to complete build.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years ago[Android] Update rootfs build script (#1476)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 31 May 2018 07:06:03 +0000 (16:06 +0900)]
[Android] Update rootfs build script (#1476)

- Install boost (header only)
- Install arm compute

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[Android] Enable build for all runtimes (#1475)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 31 May 2018 07:03:38 +0000 (16:03 +0900)]
[Android] Enable build for all runtimes (#1475)

Enable Android build for runtimes including pure_arm_compute
For link library, pthread should be parameterized.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoGeneric Tensor Comparator (#1462)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 06:54:11 +0000 (15:54 +0900)]
Generic Tensor Comparator (#1462)

This commit revises TfLiteTensorComparator as generic tensor comparator,
and move it to nnfw_util.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[PureCL] Initialize CLTensors having const data (#1472)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 06:23:03 +0000 (15:23 +0900)]
[PureCL] Initialize CLTensors having const data (#1472)

For issue #1366

In the current implementation of Pure CL runtime, CLTensors, which have
const data in their corresponding NNAPI operand, are not initialized.
This patch fixes this issue by initializing them in PlanBuilder's
finalize method.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years ago[tfliite_benchmark] Allow Thread configuration (#1473)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 06:22:47 +0000 (15:22 +0900)]
[tfliite_benchmark] Allow Thread configuration (#1473)

This commit revises tflite_benchmark to read thread configuration from
THREAD environment variable.

This commit also changes the default configuration as -1 to enable
multi-threading.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Pure CL] Allow iteration over NNAPI operands (#1471)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 06:21:59 +0000 (15:21 +0900)]
[Pure CL] Allow iteration over NNAPI operands (#1471)

This commit introduces iterate method which allows users to iterate over
registered NNAPI operands.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago [Gather Op] add other operation types (#1329)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 05:59:32 +0000 (14:59 +0900)]
 [Gather Op] add other operation types (#1329)

* [Gather Op] add uint8 operation types

This commit adds other operation types.
-uint8

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Gather Op] add int32 operation types

This commit adds other operation types.
-int32

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Gather Op] add an comment that explains gather operation

This commit adds comments.
- Gather of NN Runtime supports TENSOR_FLOAT32, TENSOR_QUANT8_ASYMM, TENSOR_INT32 other than TFLite.(TFLite additionaly supports kTfLiteString.)

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[Pure CL] Skeleton for addOperationEx (#1468)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 05:49:24 +0000 (14:49 +0900)]
[Pure CL] Skeleton for addOperationEx (#1468)

Prepare addOperationEx
It is to avoid merge conflict.

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoCode for generic tensor shape in acl/src/shape.cpp (#1404)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 05:26:22 +0000 (14:26 +0900)]
Code for generic tensor shape in acl/src/shape.cpp (#1404)

Parent issue: #1402

This code follows the explanation in https://arm-software.github.io/ComputeLibrary/latest/architecture.xhtml
```
Tensors are defined by a DataType plus a number of channels (Always expected to be 1 for now)
their dimensions are expressed as [width, height, feature_maps, batch].
In other words, the lower three dimensions of a tensor specify a single input in
[width, height, feature_maps], while any other specified dimension represents a batch
in the appropriate dimension space. For example, a tensor with dimensions [128, 128, 64, 16]
represents a 1D batch space with 16 batches of 128 elements in width and height
and 64 feature maps each.
```

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years ago[Pure CL] Extract Various Source Implementations (#1452)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 04:50:54 +0000 (13:50 +0900)]
[Pure CL] Extract Various Source Implementations (#1452)

This commit extracts various source implementations from execution.cc.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoExtract RandomGenerator from RandomTestRunner (#1445)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 31 May 2018 04:20:47 +0000 (13:20 +0900)]
Extract RandomGenerator from RandomTestRunner (#1445)

In `nnfw_util` module, extract RandomGenerator from RandomTestRunner
so RandomGenerator can be used in `tflite_run` as well.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[Pure CL Runtime] Handle shape with rank one on setInput/Output (#1449)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 02:43:37 +0000 (11:43 +0900)]
[Pure CL Runtime] Handle shape with rank one on setInput/Output (#1449)

This patch revises ANeuralNetworksExecution_setInput/Output to handle
the case of rank one.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoAdd a DIV broadcast unit test (#1459)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 02:42:38 +0000 (11:42 +0900)]
Add a DIV broadcast unit test (#1459)

For issue #1366

This commit adds a test case where DIV broadcast operation is examined.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoModify #ifndef macro of SimpleArithmeticAdditionLayer.h (#1458)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 02:42:02 +0000 (11:42 +0900)]
Modify #ifndef macro of SimpleArithmeticAdditionLayer.h (#1458)

from
`#ifndef __SIMPLE_ARITHMETIC_LAYER_H__`
to
`#ifndef __SIMPLE_ARITHMETIC_ADDITION_LAYER_H__`

because we may need more arithmetic layers.

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years ago[NNAPI test] Introduce asserts on input types (#1457)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 02:41:36 +0000 (11:41 +0900)]
[NNAPI test] Introduce asserts on input types (#1457)

This commit introduces asserts on input types as float32 is the only
input type that nnapi_test supports currently.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoDo not delegate temporary tensors to NNAPI. (#1359)
최형규/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 02:33:04 +0000 (11:33 +0900)]
Do not delegate temporary tensors to NNAPI. (#1359)

- also added delegation for MUL, and set the default scale to be 0.0f.

PiperOrigin-RevId: 195288948

This is a cherrypick of https://github.sec.samsung.net/RS7-STAR/Tensorflow/pull/41
with modification to (1) resolve conflict and (2) apply new format.

Therefore this is a modification of cherrypick of origitnal Tensorflow from
https://github.com/tensorflow/tensorflow/commit/a16ba4fc0d3faec077c689f3f361264978a2d3cb#diff-f4a5031c0f2ec4e55c3a89bf4c21c7f8

Signed-off-by: Hyung-Kyu Choi <hk0110.choi@samsung.com>
6 years agoCAST_EX nnapi_delegate (#1429)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 02:28:49 +0000 (11:28 +0900)]
CAST_EX nnapi_delegate (#1429)

1. Add add_cast_ex_params
2. Add nnapi call from nnapi_delagate.cpp

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoBuild Android TFLite JNI shared library (#1442)
박세희/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 01:24:49 +0000 (10:24 +0900)]
Build Android TFLite JNI shared library (#1442)

* Build Android TFLite JNI shared library

This will change cmake to drop Android JNI including Tensorflow Lite
- current drop will work same as vanilla TFLite library

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

6 years agoFix linking error on arm compute library (#1456)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Thu, 31 May 2018 01:24:25 +0000 (10:24 +0900)]
Fix linking error on arm compute library (#1456)

Recently introduced `ARMCompute` removed link to dl and pthread.
So it produces linking error like following:

```
arm-linux-gnueabihf/bin/ld:
warning: libdl.so.2, needed by
/home/brian/z/nnfw/Product/armv7l-linux.debug/out/lib/libarm_compute.so,
not found (try using -rpath or -rpath-link)
/home/brian/z/nnfw/Product/armv7l-linux.debug/out/lib/libarm_compute_core.so:
undefined reference to `dlopen@GLIBC_2.4'
/home/brian/z/nnfw/Product/armv7l-linux.debug/out/lib/libarm_compute_core.so:
undefined reference to `dlsym@GLIBC_2.4'
/home/brian/z/nnfw/Product/armv7l-linux.debug/out/lib/libarm_compute_core.so:
undefined reference to `dlerror@GLIBC_2.4'
collect2: error: ld returned 1 exit status
```

By specifying dl and pthread as dependency of arm_compute_xxx, the
linking error could be fixed.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoUpdate ACL submodule hash (#1454)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 01:21:49 +0000 (10:21 +0900)]
Update ACL submodule hash (#1454)

This patch updates the submodule hash for ACL to the latest one.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoUse submodule on tizen cross build (#1437)
이춘석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 01:05:56 +0000 (10:05 +0900)]
Use submodule on tizen cross build (#1437)

* Use submodule on tizen cross build

1. Let OBS_BUILD have ON/OFF
2. Use submodule on tizen cross build

In summary, tizen cross build works in the same way of arm cross build.

* Change submodule during tizen cross build

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoEnable test_driver to catch failed test (#1409)
김용섭/동작제어Lab(SR)/Engineer/삼성전자 [Thu, 31 May 2018 00:59:02 +0000 (09:59 +0900)]
Enable test_driver to catch failed test (#1409)

* Enable test_driver to catch failed test

Enable tools/test_driver/test_driver.sh to catch failed test, so that
now do not pass any failed test cases. To do this, run_benchmark*.sh and
run_unittest.sh will stop when the tests catch fail.
run_frameworktest.sh will stop after the tests run all models.

Signed-off-by: Yongseop Kim <yons.kim@samsung.com>
6 years ago[Pure CL] Support Tensor Sinks (#1440)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 00:56:31 +0000 (09:56 +0900)]
[Pure CL] Support Tensor Sinks (#1440)

This commit revises pure CL runtime to support generic tensor outputs.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoMake iterate() static inline (#1451)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Thu, 31 May 2018 00:47:06 +0000 (09:47 +0900)]
Make iterate() static inline (#1451)

This patch makes iterate() static inline function in order to avoid
multiple definition issue that could be caused when IndexIterator.h is
included in multiple files.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoAdd hasData() to class Object (#1450)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 23:57:33 +0000 (08:57 +0900)]
Add hasData() to class Object (#1450)

Object::hasData() will be used to check if Object has data internally.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoAdd a DIV unit test (#1448)
서상민/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 23:57:16 +0000 (08:57 +0900)]
Add a DIV unit test (#1448)

This patch adds a DIV unit test where two tensors with the same same are
passed as inputs.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoIntroduce 'tf_test' tool (#1421)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 23:41:31 +0000 (08:41 +0900)]
Introduce 'tf_test' tool (#1421)

This commit introduce experimental 'tf_test' tool. This tool loads
tensorflow lite model and corresponding tensorflow model, run
inferences, and compare the result.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoTensorFlowMax nnapi_delegate (#1432)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 23:41:01 +0000 (08:41 +0900)]
TensorFlowMax nnapi_delegate (#1432)

Add nnapi call from nnapi_delegate.cpp for TensorFlowMax

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoTFLite interpreter: Custom operator max (#1419)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 22:24:51 +0000 (07:24 +0900)]
TFLite interpreter: Custom operator max (#1419)

General implementation for custom operator max
Similar with mean

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoOperator selector: cast operator bug fix (#1428)
오형석/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 22:23:38 +0000 (07:23 +0900)]
Operator selector: cast operator bug fix (#1428)

Fix bug about generating cast operator's option

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[newrt] Apply fix to newrt (#1444)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 30 May 2018 22:23:15 +0000 (07:23 +0900)]
[newrt] Apply fix to newrt (#1444)

Apply fixed that were applied to pure_arm_compute

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[Topk_v2 Op] support other operation types (#1314)
최성진/동작제어Lab(SR)/Principal Engineer/삼성전자 [Wed, 30 May 2018 11:06:45 +0000 (20:06 +0900)]
[Topk_v2 Op] support other operation types (#1314)

* [Topk_v2 Op] add uint8 types

This commit adds other types.
-uint8

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Topk_v2 Op] add int32 operation types

This commit adds other types.
-int32

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Topk_v2 Op] add an comment that explains topk_v2 operation

This commit adds comments.
- TopK_v2 of NN Runtime supports TENSOR_FLOAT32, TENSOR_QUANT8_ASYMM, TENSOR_INT32 other than TFLite.

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
* [Top[Topk_v2 Op] add 2D test cases for other types

This commit adds 2D test cases for other types (int32, uint8)

Signed-off-by: SungJin Choi <lotieye.choi@samsung.com>
6 years ago[Pure CL] Inheric nnfw::utils::tensor::Shape (#1438)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 10:53:59 +0000 (19:53 +0900)]
[Pure CL] Inheric nnfw::utils::tensor::Shape (#1438)

This commit revises internal::tflite::operand::Shape to inherit
nnfw::utils::tensor::Shape (to reduce code duplication).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Pure CL runtime] Support reshape/softmax over tensors (#1436)
박종현/동작제어Lab(SR)/Staff Engineer/삼성전자 [Wed, 30 May 2018 08:49:00 +0000 (17:49 +0900)]
[Pure CL runtime] Support reshape/softmax over tensors (#1436)

This commit removes asserts in reshape/softmax compilation
implementation to support ADAS model.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoIntroduce 'tensor::Diff' class (#1422)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 08:04:43 +0000 (17:04 +0900)]
Introduce 'tensor::Diff' class (#1422)

This commit extracts general 'tensor::Diff' clss from
'TfLiteTensorDiff'.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoFix format for pure acl runtime (#1435)
오형석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 07:53:27 +0000 (16:53 +0900)]
Fix format for pure acl runtime (#1435)

Fix format for pure acl runtime
Enable format check for pure acl runtime

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoIgnore compiler warnings from ARMCompute (#1434)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 07:47:00 +0000 (16:47 +0900)]
Ignore compiler warnings from ARMCompute (#1434)

This commit adds -Wno-ignored-attributes option when building pure CL
runtime to ignore compiler warnings due to the code in ARMCompute
library (CL.h).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoFix build break on debug build (#1427)
이춘석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 07:31:16 +0000 (16:31 +0900)]
Fix build break on debug build (#1427)

Fix debug build break for pure acl runtime

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoUpdate TensorFlow submodule for topk_v2 (#1420)
오형석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 06:21:44 +0000 (15:21 +0900)]
Update TensorFlow submodule for topk_v2 (#1420)

This patch updates the submodule hash for tensorflow to handle topk_v2

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years ago[Pure CL runtime] Fix CMakeLists.txt indentation (#1424)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 05:36:53 +0000 (14:36 +0900)]
[Pure CL runtime] Fix CMakeLists.txt indentation (#1424)

This commit fixes CMakeLists.txt indentation.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoAdd 'TFLITE_ENSURE' macro (#1418)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 02:25:51 +0000 (11:25 +0900)]
Add 'TFLITE_ENSURE' macro (#1418)

This commit introduces 'TFLITE_ENSURE' macro, and use it for tflite_run
and tflite_benchmark.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Pure CL Runtime] Use SimpleArithmeticAdditionLayer (#1414)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 02:23:54 +0000 (11:23 +0900)]
[Pure CL Runtime] Use SimpleArithmeticAdditionLayer (#1414)

This commit introduces SimpleArithmeticAddationLayer, and uses it by
default kernel on pure CL runtime.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoIntroduce 'ARMCompute' package (#1400)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 02:20:56 +0000 (11:20 +0900)]
Introduce 'ARMCompute' package (#1400)

This commit introduces 'ARMCompute' package which provides uniform
interface over imported/built ARMCompute libraries.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoUpdate xu4 documentations to more detail about minicom (#1334)
김수진/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 30 May 2018 02:12:32 +0000 (11:12 +0900)]
Update xu4 documentations to more detail about minicom (#1334)

* Update xu4 documentations for using keyboard control

This commit updates xu4 documentations to more detail for using keyboard control

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

* Change comments to more cleary

6 years agoEnable Android build of tflite_benchmark, disable tflite_run (#1413)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Wed, 30 May 2018 01:32:10 +0000 (10:32 +0900)]
Enable Android build of tflite_benchmark, disable tflite_run (#1413)

Enable/Disable build for Android

- Enable build of tflite_benchmark
- Disable tflite_run since it has boost library build issue
  (Related to standard library - gnustl/libc++)

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoSupport Strided Slice Ex on nnapi_delegate.cc (#1399)
이춘석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Wed, 30 May 2018 01:30:28 +0000 (10:30 +0900)]
Support Strided Slice Ex on nnapi_delegate.cc (#1399)

* Support Strided Slice Ex on nnapi_delegate.cc

Add Strided Slice Ex conversion to nnapi_delegate.cc
Note that stride_slide on nn runtime does not support 2 parameters : ellipsis_mask
and new_axis_mask.

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* add CHECK_NN, remove break

6 years agoRemove CXX_STANDARD workaround (#1405)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Tue, 29 May 2018 22:33:08 +0000 (07:33 +0900)]
Remove CXX_STANDARD workaround (#1405)

* Remove CXX_STANDARD workaround

CXX_STANDARD is no longer necessary as the minimum required version of
CMake is 3.1 (which supports CXX_STANDARD feature).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
* Apply the same change to 'libs/kernel/acl/CMakeLists.txt'

6 years ago[Pure CL runtime] Add symbolcheck test (#1397)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Tue, 29 May 2018 11:09:00 +0000 (20:09 +0900)]
[Pure CL runtime] Add symbolcheck test (#1397)

This commit introduces 'pure_arm_compute_symbolcheck' executable which
allows us to detect undefined references in pure CL runtimes during
build.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoRevert "Update validation in runtime (#1351)" (#1406)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Tue, 29 May 2018 05:51:07 +0000 (14:51 +0900)]
Revert "Update validation in runtime (#1351)" (#1406)

This reverts commit 86bbecea1fce90cac87a40782900eb24ff09db95.

PR #1351 caused many test failures, which were not noticed by our CI.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoUpdate validation in runtime (#1351)
장지섭/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 29 May 2018 02:01:53 +0000 (11:01 +0900)]
Update validation in runtime (#1351)

This commit updates validation in runtime.

Signed-off-by: jiseob.jang <jiseob.jang@samsung.com>
6 years agoInstall arm_compute_XXX and libnueralnetworks.so on pure_arm_compute … (#1393)
이춘석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Tue, 29 May 2018 01:25:00 +0000 (10:25 +0900)]
Install arm_compute_XXX and libnueralnetworks.so on pure_arm_compute … (#1393)

* Install arm_compute_XXX and libnueralnetworks.so on pure_arm_compute build

1. On pure_arm_compute runtime build, install arm_compute_XXX.so into lib
2. DO NOT build NN Runtime if BUILD_ACL=ON

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* Add option BUILD_PURE_ARM_COMPUTE and change install dir

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
* Remove `BUILD_NN_RUNTIME` option off

Since pure_acl_compute runtime is installed to lib/pureacl, we do not need turn
off option BUILD_NN_RUNTIME.

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

In summary,
1. install arm_compute_xxx to lib
2. install libnn.so to lib/pureacl

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoConvert TF Lite's DIV to NN API's (#1398)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Tue, 29 May 2018 01:22:40 +0000 (10:22 +0900)]
Convert TF Lite's DIV to NN API's (#1398)

For #1199 and #1366

This commit links TF Lite's DIV operation to NN API's.  It also enables
the Div test in the framework test.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years ago[pure_arm_compute] Fix no return value (Context &) (#1396)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Tue, 29 May 2018 00:47:33 +0000 (09:47 +0900)]
[pure_arm_compute] Fix no return value (Context &) (#1396)

`Context::set()` is non-void but did not return a value.
Make it to return `*this`.

Similar to #1382

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years agoEnable SUB NNAPI delegation and SUB broadcast test (#1394)
이상규/동작제어Lab(SR)/Principal Engineer/삼성전자 [Tue, 29 May 2018 00:47:11 +0000 (09:47 +0900)]
Enable SUB NNAPI delegation and SUB broadcast test (#1394)

Tensorflow is updated to v.1.8 so we can run SUB broadcast.
I pave the way to NN runtime for Sub, (It's same to ADD.)
and enabled the corresponding test.

Signed-off-by: Sanggyu Lee <sg5.lee@samsung.com>
6 years agoIntroduce 'Tensorflow' package (#1387)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 11:13:50 +0000 (20:13 +0900)]
Introduce 'Tensorflow' package (#1387)

This commit introduces 'Tensorflow' package which allows users to use
externally built libtensorflow-core.a as if CMake targets.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoMake Tensorflow Lite calls MUL CPU fallback (#1360)
윤현식/동작제어Lab(SR)/Principal Engineer/삼성전자 [Mon, 28 May 2018 10:43:11 +0000 (19:43 +0900)]
Make Tensorflow Lite calls MUL CPU fallback (#1360)

1. Modified `NNAPIDelegate` to build graph with `MUL`
2. Test with ADAS 'MUL' graph by nnapi_test

Signed-off-by: Hyun Sik Yoon <hyunsik.yoon@samsung.com>
6 years agoUpdate TensorFlow submodule to v1.8 (#1388)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 08:32:50 +0000 (17:32 +0900)]
Update TensorFlow submodule to v1.8 (#1388)

For #1383

This patch updates the submodule hash for TensorFlow to v1.8 branch in
RS7-STAR/Tensorflow.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoInclude model.h in nnapi_delegate.cpp (#1386)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 07:27:21 +0000 (16:27 +0900)]
Include model.h in nnapi_delegate.cpp (#1386)

For #1383

Although TF Lite's `nnapi_delegate.cc` includes
`tensorflow/contrib/lite/model.h`, our nnapi_delegate.cpp did not
include `model.h` because we were able to compile `nnapi_delegate.cpp`
without the header file.  However, it is not the case with the latest
TensorFlow v1.8, which we will be using soon.  This patch is to prevent
the build failure that could be caused by not including `model.h` when
building with TensorFlow v1.8.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoIntroduce nnfw's BuiltinOpResolver (#1384)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 07:19:56 +0000 (16:19 +0900)]
Introduce nnfw's BuiltinOpResolver (#1384)

For issue #1383

This patch introduces our own BuiltinOpResolver, which is basically
copied from externals/tensorflow, in order to manage which builtin
operators in TF Lite are used with nnfw.

To only use nnfw's BuiltinOpResolver, this patch also excludes
register.cc in externals/tensorflow from the TF Lite build.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years ago[pure_arm_compute] Fix no return value (#1382)
이한종/동작제어Lab(SR)/Engineer/삼성전자 [Mon, 28 May 2018 07:09:16 +0000 (16:09 +0900)]
[pure_arm_compute] Fix no return value (#1382)

`Sequence::emplace_back()` is non-void but did not return a variable.

Signed-off-by: Hanjoung Lee <hanjoung.lee@samsung.com>
6 years ago[Pure CL runtime] Fix undefined references (#1380)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 06:31:22 +0000 (15:31 +0900)]
[Pure CL runtime] Fix undefined references (#1380)

This commit fixes undefined reference issues on pure CL runtime.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoInclude cstring for memcpy (#1378)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 05:35:42 +0000 (14:35 +0900)]
Include cstring for memcpy (#1378)

This patch adds `#include <cstring>` to tensor_dumper.cc, since `memcpy`
is used in the file.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoAdd Extended NNAPI shim header (#1365)
이춘석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 05:35:23 +0000 (14:35 +0900)]
Add Extended NNAPI shim header (#1365)

* Add Extended NNAPI shim header

Add Extended NNAPI shim header. This header includes
ANEURALNETWORKS_STRIDED_SLICE_EX = 50005

Signed-off-by: Chunseok Lee <chunseok.lee@samsung.com>
6 years agoInclude cassert in Diff.cpp (#1377)
서상민/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:33:05 +0000 (13:33 +0900)]
Include cassert in Diff.cpp (#1377)

This patch adds `#include <cassert>` to Diff.cpp, since `assert()` is
used in the file.

Signed-off-by: Sangmin Seo <sangmin7.seo@samsung.com>
6 years agoIntroduce 'nnfw_find_package' (#1372)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:32:49 +0000 (13:32 +0900)]
Introduce 'nnfw_find_package' (#1372)

This commit introduces 'nnfw_find_package' macro which allows users to
write and use nnfw-internal CMake modules (similarly as modules provided
via find_package).

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoUse nnfw-internal NNAPIDelegate (#1370)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:11:38 +0000 (13:11 +0900)]
Use nnfw-internal NNAPIDelegate (#1370)

This commit revises NNAPISession to use nnfw-internal NNAPIDelegate.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoAdd tensorflow-based 'detection' example (#1368)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:10:33 +0000 (13:10 +0900)]
Add tensorflow-based 'detection' example (#1368)

This commit adds 'detection' sample app which loads T/F model (which
takes 1x320x320x3 input whose name is 'input_node') and measures elapsed
time for 5 runs.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoUse common benchmark framework (#1371)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:07:48 +0000 (13:07 +0900)]
Use common benchmark framework (#1371)

Thsi commit revises tflite_run and tflite_benchmark to use common
benchmark framework.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoDo NOT install libtensorflow-lite.a (#1375)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 04:06:45 +0000 (13:06 +0900)]
Do NOT install libtensorflow-lite.a (#1375)

libtensrflow-lite.a is used only during nnfw build.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years ago[Pure CL runtime] Support general reshape (#1353)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 02:39:55 +0000 (11:39 +0900)]
[Pure CL runtime] Support general reshape (#1353)

* [Pure CL runtime] Support general reshape

This commit removes height/width restriction on reshape operation in
pure CL runtime.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoMove to CMake 3.1 (#1369)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 02:28:51 +0000 (11:28 +0900)]
Move to CMake 3.1 (#1369)

This commit requires CMake 3.1 as minimum required CMake version for
nnfw.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoOperator selector: support topk_v2 and gather (#1361)
오형석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Mon, 28 May 2018 01:35:57 +0000 (10:35 +0900)]
Operator selector: support topk_v2 and gather (#1361)

Support topk_v2 and gather operator

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agoShow every tensor diffs (#1356)
박종현/동작제어Lab(SR)/Senior Engineer/삼성전자 [Fri, 25 May 2018 10:17:49 +0000 (19:17 +0900)]
Show every tensor diffs (#1356)

* Show every tensor diffs

This commit revises TfLiteInterpMatchApp to show all the tensor matching
results even when there are unmatched tensors.

Signed-off-by: Jonghyun Park <jh1302.park@samsung.com>
6 years agoAdd topk_v2 and gather framework test (#1362)
오형석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Fri, 25 May 2018 09:18:08 +0000 (18:18 +0900)]
Add topk_v2 and gather framework test (#1362)

Add topk_v2 and gather framework test

Signed-off-by: Hyeongseok Oh <hseok82.oh@samsung.com>
6 years agomake extern C for extended NNAPI (#1336)
이춘석/동작제어Lab(SR)/Senior Engineer/삼성전자 [Fri, 25 May 2018 08:55:32 +0000 (17:55 +0900)]
make extern C for extended NNAPI (#1336)

* make extern C for extended NNAPI

Add __BEGIN_DECLS for extended NNAPI

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