Vladimir Paramuzov
a73fc2dce1
[GPU] Fix strided_slice for dynamic cases ( #12979 )
2022-09-13 09:17:50 +04:00
Ilya Lavrenov
3c24ee6cda
Install tbbmalloc for pkg-config case ( #12960 )
2022-09-12 22:12:50 +04:00
Andrii Pavliuk
69f709028a
[benhmark_app] Fix I8 input random generation ( #12502 )
2022-09-12 15:21:40 +03:00
Przemyslaw Wysocki
006f798f91
[PyOV] Fix numpy deprecation warnings ( #12849 )
2022-09-12 12:32:23 +02:00
Oleksii Khovan
0145865301
[GPU] Softmax blocked layouts support ( #12467 )
2022-09-12 13:48:45 +04:00
Vladimir Paramuzov
621bf375c1
[GPU] Minor fix for dynamic reshape ( #12989 )
2022-09-12 13:45:49 +04:00
Jan Iwaszkiewicz
8fae2ddfe3
update docs ( #12974 )
2022-09-12 11:04:54 +02:00
Artur Kulikowski
1b620fa8bc
Add access to the methods Node::evaluate to each custom Op which inherit from ov::op::Op ( #12976 )
...
* Add access to the hidden Node::evaluate method in each Ops
* Fix the test
* Add new line in EOF
* Add comment about using ov::op::Op
* Use opset9
* Add more detailed comment
2022-09-12 07:51:16 +03:00
Helena Kloosterman
49ebb95067
Typo fixes in main README ( #12997 )
2022-09-12 08:20:04 +04:00
Ilya Lavrenov
bc8127f3b8
Fixed Android post-commit ( #13005 )
2022-09-12 07:57:21 +04:00
Artur Kulikowski
c792db8302
VaridicSplit ref impl: No evaluate output if has zeros dimension ( #12939 )
...
* VaridicSplit ref impl: No evaluate output if has zeros dimension
* Move check of zeros dims in output shape to the variadic_split::evaluate method
2022-09-12 07:34:37 +04:00
Ekaterina Aidova
d4d9bb4e88
[master] openvino-dev return opencv-python back ( #13002 )
2022-09-10 12:18:03 +04:00
Ilya Lavrenov
53542ded57
Added support of pugixml search via pkg-config ( #12958 )
...
* Added support of pugixml search via pkg-config
* Updated old InferencEngineDEveloperPackage.cmake.in
* Removed debug print
* Fixed cmake stage for U18
* Added WA for pugixml on U18 for DevPackage as well
* Disable pkg-config generation for OpenVINO for U18
* Added option for pkg-config
* Now triplet is aut-detected
* Fixed U18 case
2022-09-10 01:10:24 +03:00
Ilya Lavrenov
ee8575134a
enabled LTO for GPU plugun helper libs ( #12891 )
2022-09-10 00:58:28 +03:00
Vladimir Paramuzov
1ded0750c9
[GPU] Remove marks cleanup after each pass ( #12975 )
2022-09-09 18:29:31 +03:00
Xuejun Zhai
349f3e02e0
[C API] Add more C API for infer request ( #12636 )
...
* [C API] Add more C API for infer request
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [Code Style] Fix code style issue
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* Fix review comments
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API] add more details desc for infer request interfaces
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API] Modify the structure of tensor list
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API] Fix code style issue
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API] remove define about tensor list
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* Update documents for infer request
Change-Id: I1acbae95425cfe4aad85da48d290bafe5bb7d319
* Update documents for ov_infer_request
Change-Id: Ib68530c49cf439c7b570ee63bd3ee8ea85018d51
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
Co-authored-by: River,Li <river.li@intel.com>
2022-09-09 10:47:46 +03:00
Gorokhov Dmitriy
e7fe00f5f2
[CPU] Support MHA optimization ( #12936 )
2022-09-09 10:03:19 +04:00
yanlan song
0dd1f6e1b0
Bell/wa multi perf gap ( #12948 )
...
* debug only
Signed-off-by: fishbell <bell.song@intel.com>
* refine comment
Signed-off-by: fishbell <bell.song@intel.com>
* Revert "debug only"
This reverts commit 5ef6140b63
2022-09-09 10:14:00 +08:00
mei, yang
b2d37ff85a
add GenerateProposals single layer test ( #12933 )
...
Co-authored-by: Andrei Kochin <andrei.kochin@intel.com>
2022-09-08 17:35:14 +03:00
Mateusz Tabaka
7b98ebf627
[ONNX] Add support for Trilu operator ( #12965 )
...
Ticket: 91096
2022-09-08 12:51:28 +02:00
Mykhailo Hnap
e0f6cdaa85
Added better exception handling to op_imp_check. ( #12961 )
2022-09-08 10:44:38 +03:00
Andrew Kwangwoong Park
90093d5315
[GPU] Fix the new infer request to set output memory normally during enqueue stage ( #12938 )
...
* Fix the new infer request to set output memory normally during enqueue stage
* Fix to use the generated outputsMap instead of graph API to retrieve outputID
Signed-off-by: Andrew Park <andrew.park@intel.com>
2022-09-08 16:15:04 +09:00
Tomasz Jankowski
5a42afb45b
[Core] Reduce unit test dependencies ( #12932 )
...
* Remove manifest dependency
* Use generic TEST macro
* Remove test_control dependency
* Redo Sqrt layer test
* Remove unused files
Co-authored-by: Ilya Churaev <ilya.churaev@intel.com>
2022-09-08 09:04:04 +03:00
Roman Kazantsev
0b1a70be0b
[TF FE] Add translators for NormalizeL2, ReverseSequence ( #12913 )
...
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2022-09-07 18:02:48 +03:00
Ilya Lavrenov
8922d73e7d
Fixed installation of pugixml on U22 ( #12953 )
2022-09-07 18:07:43 +04:00
Yuan Xu
67d5652b13
Fix a link anchor on pypi page ( #12949 )
...
* fix the link for pip
* update the <a> tags
2022-09-07 18:07:22 +04:00
Ilya Churaev
97c510c008
Fixed build for TBB which uses pre-release functions ( #12947 )
...
* Fixed build for TBB which uses pre-reliase functions
* Disable TBB only for macOS only
* Changed condition
2022-09-07 18:05:41 +04:00
River Li
b23ac86097
tbbbind support CC ( #12107 )
...
* Add CC support for tbbbind
* Use throw to replace return
* Tbbbinder will be optimized out for non-NUMA
1. With NUMA we may expect that some TBBBind API is optimized out, but some of the API is definitely here
2. The same for HYBRID cores on Alder lake
3. For no NUMA nor HYBRID I expect no TBBBind API is used
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
2022-09-07 11:14:05 +04:00
Pawel Raasz
2449ab066a
Delete pybind11 cross compile workaround ( #12775 )
...
Co-authored-by: sikorsl1 <leonard.sikorski@intel.com>
2022-09-07 10:54:11 +04:00
Yuan Xu
17470aaea9
fix the link for pip ( #12945 )
2022-09-07 10:17:46 +04:00
Fedor Zharinov
4b727de6da
[benchmark_app] Separate hint from fine tuning options. ( #10919 )
...
* Separate hint from fine tuning options.
* stylefix
* typo fix
* Order is changed
2022-09-07 10:13:34 +04:00
RICKIE777
f780303e5a
Fix the memory issue in ov_core_read_model_from_memory ( #12817 )
2022-09-07 10:06:31 +04:00
Ilya Lavrenov
a092a192d8
Ability to build against oneTBB build RelWithDebInfo ( #12940 )
...
* Ability to build against oneTBB build RelWithDebInfo
* Debug TBB issue
* Fixes
2022-09-07 09:58:30 +04:00
River Li
4df8b279c8
Unify unchanged input arguments as const arguments ( #12915 )
...
Change-Id: Icd9f369a72b02ff35cd4a239bddf4c24b6c2af29
2022-09-07 09:52:58 +04:00
Tetiana Gubanova
8979dc562d
[GPU] Fix implementation map includes to fix the dependency ( #12043 )
...
* Reorder includes to check that build failed
* Add needed includes to remove dependency on include orders in primitives
2022-09-06 22:21:21 -07:00
Tetiana Gubanova
cfbe141b0c
Add blocked layouts support to bucketize ( #12079 )
2022-09-06 22:20:14 -07:00
Oleksii Khovan
c68593e03b
[GPU] ScatterUpdate blocked layout support ( #12639 )
...
* blocked formats support
* tests for blocked formats
* review comments
* fix RHEL 8.2 build
2022-09-07 08:36:43 +04:00
Vladimir Paramuzov
734dcc93e9
[GPU] Use oneDNN gemm on DG2 in some cases ( #12878 )
...
* [GPU] Enable onednn gemm with some additional restrictions
* f64 support in c++ benchmark_app
* Fixed python benchmark_app
2022-09-07 08:35:14 +04:00
Jade Cho
d8b6f5485b
[GPU] Fix a bug of graph optimization ( #12901 )
...
+ Use program_node to find weight node of convolution at eltwise shrinking
2022-09-07 10:43:30 +09:00
Nikita Malinin
ace759c74c
Cherry-pick changes from 6d69a85 ( #12931 )
...
Co-authored-by: Indira Salyahova <indira.salyahova@intel.com>
2022-09-06 17:45:46 +00:00
Vladislav Golubev
88e4ac5e53
[LPT] Fixes for the cases with convert before subtraction constant ( #12835 )
2022-09-06 20:41:29 +03:00
Ilya Lavrenov
9d55355daf
OpenVINO requires TBB 2017.0 or higher ( #12934 )
...
* Added a error message if cmake < 3.16 is used with tbb.pc
* TBB
* Updates
* Ignore old TBBs
* Handle minimal version via pkg-config as well
2022-09-06 18:57:55 +04:00
Liubov Talamanova
53319e60fc
Fixed insert_fake_quantize() with empty hw_config ( #12676 )
2022-09-06 17:48:28 +04:00
Ilya Lavrenov
10f2cf40eb
Cross-compilation of debian packages ( #12919 )
...
* Debian arm cross-compilation
* Pugixml::shared as system library
* Removed debug print of openvino.pc
* Set debian package arch for cross-compilation manually
2022-09-06 12:03:40 +03:00
Adam Tumialis
ab1da94f1d
Update CODEOWNERS ( #12723 )
...
* Update CODEOWNERS
New scheme for CODEOWNERS file.
* Update CODEOWNERS
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
* Update CODEOWNERS
Reverted changes to /docs/template_plugin/ and /docs/template_extension
* Update CODEOWNERS
Reverted changes for GPU & VPU plugin include paths.
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
2022-09-06 11:00:12 +02:00
Mateusz Bencer
2076aa8fc7
added op check tests for RDFT and IRDFT ( #12917 )
2022-09-06 11:52:52 +03:00
Yuan Xu
d6255baae8
update system requirements & pre-release disclaimer ( #12717 )
...
* update system requirements & pre-release disclaimer
* change latest to nightly documentation
* update C++ requirements according to comments
2022-09-06 12:52:08 +04:00
Ilya Lavrenov
4d8d119dc5
Added a error message if cmake < 3.16 is used with tbb.pc ( #12914 )
2022-09-06 12:03:32 +04:00
Felix Dohyun Kim
0b05eb2c67
[GPU] Remove scale primitive ( #12893 )
2022-09-06 11:35:26 +04:00
Felix Dohyun Kim
2108fe0dfc
[GPU] Run onednn test only in DG2 ( #12892 )
2022-09-06 09:02:15 +09:00
