Commit Graph

6972 Commits

Author SHA1 Message Date
guozhong wang
2d88e67616
Guozhong/remove format time (#9923)
* remove formatTimeMilli from time_utils.cpp

* add traceCallStacks test case

* add traceCallStacks test case in format_test.cpp

* add param:"test" to function TraceCallStacks()

* rollback file in master branch

* add traceCallStacks test case in format_test.cpp

* remove formatTimeMilli from time_utils.cpp and add traceCallStacks test case in format_test.cpp
2022-02-18 09:36:01 +08:00
Mingyu Kim
10ac5b280b
[GPU] Mixed precision fix for mask rcnn (#10467)
* Select proper layout for fp16-int8 mixed precision network
* Set proper layout in layout propagation for mixed precision
2022-02-18 10:27:54 +09:00
hyunback kim
215db2dad8
[GPU] Enable shuffle and fsv32 in implicit concat (#9888)
[GPU] Enable shuffle and fsv32 in implicit concat

* Support shuffle fsv32
* Check feature depths in first input depedency.
* Add to select onednn convolution in case block format in get_preferred_impl_type func.

Signed-off-by: hyunback <hyunback.kim@intel.com>
2022-02-18 09:40:14 +09:00
Vladislav Volkov
b6a75d7d91
CPU plugin namespaces and files renaming (#10248) 2022-02-17 23:48:26 +03:00
Nikolay Tyukaev
7fa9d07a1f
ignore api/reference.rst warnings (#10491) 2022-02-17 23:13:09 +03:00
Maxim Vafin
f482f9765e
Fix values reading from protobuf (#10391)
* Fix values reading from protobuf

* Remove exception

* Small fix
2022-02-17 21:42:49 +03:00
Maxim Vafin
ac880f601c
Fix getting attribute in TF FE (#10458)
* if attribute is not present and default value provided it should return default value
2022-02-17 20:30:04 +03:00
Anton Grishin
b8bbe056b1
Improve l-capturing (#10468) 2022-02-17 19:17:07 +03:00
Artur Kulikowski
73caba0f67
Fix ONNX boolean tests (#10404) 2022-02-17 15:31:11 +00:00
Nikita Malinin
a090abbc92
Update remove_converts pass with shape inference (#10474) 2022-02-17 18:17:07 +03:00
Yegor Kruglov
6e5eb87340
Add note to YOLO-v3 conversion instructions (#10428)
* added note to yolo v3 conversion instructions

* fix typo

* Update docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_YOLO_From_Tensorflow.md

style fix

Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>

Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>
2022-02-17 18:00:38 +03:00
Ivan Tikhonov
ade4c6c7f9
OpExtension: fix framework attributes handling (#10445)
* Fix attribute handling in OpExtension, add unit tests

* add missed file

* fix warning

* fix warning

* rename convert_from_py_object method to py_object_to_any, fix PEP8

* fix PEP8

* delete redundant include dir, fix includes
2022-02-17 17:42:12 +03:00
Anton Pankratov
61f657795c
Streams property with special values (#10411)
* Streams  property with special values

* Fixed clang
2022-02-17 16:39:06 +03:00
Fedor Zharinov
198f44fdc7
Fix for missing throughput in case of Multi device (#10407)
* Fix for missing throughput in case of Multi device

* stylefix
2022-02-17 16:32:19 +03:00
Ilya Lavrenov
306b7611d9
repair TF FE tests after build (#10432)
* repair TF FE tests after build

* Small improvements

* Fixed static build
2022-02-17 16:28:24 +03:00
Maxim Gordeev
3144c5fab8
Added processing of layout for speech sample (#10254)
* Added processing of layout for speech sample

* fixed notes

* some improvements

* Code style format

* changed NCC value for NullStatement

* improved batch processing

* added loading batch for imported model

* fixed notes

* fixed notes

* added layout parameter to azure tests
2022-02-17 16:11:57 +03:00
Irina Efode
ccd7104108
[IE TESTS][CONFORMANCE] Add support of dynamic shapes in SubgraphDumper (#10380)
* Initial commit. Need to remove debug code

* Remove extra flags. Fix comparation in the matchers

* Fix small issue with the default args

* Update eltwise.hpp

* Update ov_subgraph.cpp
2022-02-17 15:52:37 +03:00
Nikolay Tyukaev
fc1157cf68
add api folder if enable python (#10477) 2022-02-17 15:24:29 +03:00
Egor Shulman
8ae4bc95fd
[CPU] Coverity fixes (#10392) 2022-02-17 15:11:18 +03:00
Anton Pankratov
0882f863d6
Any compilation time optimization (#10335)
* Optimized any compilation time

* Fixed Any  compilation time

* any::addressof

* reverted

* Fixed read write

* format fix

* Fixed build

* format fix

* Moved any tests back

* removed inline

* fix format

* used static inline

* format fix

* removed inline static

* fixed merge confkicts
2022-02-17 14:55:37 +03:00
Anton Pankratov
7ce9801ec3
Added mkldnn ov properties test for compile_model (#10442)
* Added mkldnn ov properties test

* fixed  macos build
2022-02-17 13:54:02 +03:00
Anton Pankratov
d1378d94b8
Fixed default inference precision in benchmark app (#10443) 2022-02-17 13:53:50 +03:00
Vladislav Golubev
ff4e97ab09
[LPT] Security fixes (#10465) 2022-02-17 13:47:27 +03:00
Anton Chetverikov
e444715c8d
[MO] Restore inputs order in IR Reader (#10403)
* Restore inputs order in IR Reader

* Add saving of outputs order
2022-02-17 13:07:34 +03:00
Tomasz Dołbniak
83a8ac800c
ONNX model validator enhancements (#10456) 2022-02-17 11:01:47 +01:00
Anton Voronov
61f915b4f6
[CPU] changed checks with_cpu_x86...() to mayiuse() (#9911) 2022-02-17 12:56:55 +03:00
Pavel Esir
43784e2cec
fix convert_nms_gather_path_to_unsigned: added opset8::Slice into patter_list (#10439) 2022-02-17 12:47:25 +03:00
Aleksandr Korolev
8abb949af9
[VPU] Coverity fixes (#10396)
Tickets:
-79244
-78866
2022-02-17 12:29:28 +03:00
Aleksandr Korolev
5ace7bb96f
[MYX] Added missing supported properties in GetMetric method (#10440) 2022-02-17 12:23:41 +03:00
Anton Pankratov
a7b28953e2
Added Import export device capability into hetero plugin (#10455) 2022-02-17 12:15:45 +03:00
hyunback kim
8148921fa7
[GPU] Fix deconv b32 onednn regression in onednn (#10462)
After enabling deconv b32 onednn, colorization-siggraph f16 b32 has regresison,
Fix it. Add to check sum post ops in case deconv onednn.

Signed-off-by: hyunback <hyunback.kim@intel.com>
2022-02-17 18:09:51 +09:00
Irina Efode
68f523010e
[IE TESTS][CONFORMANCE] Support dynamic shapes in Operation Conformance (#10400)
* emove namespeca unity

* [IE TESTS][IE CONFORMANCE] Suppot dynamic shapes in Operation Conformance runner

* Update CMakeLists.txt

* Fix dim generation
2022-02-17 11:27:45 +03:00
hyunback kim
ed323afc93
[GPU] Remove default bfyx quantize in get_preferred_format (#9654)
* [GPU] Remove default bfyx quantize in get_preferred_format

Default bfyx occurs redundant reorder in fsv-format network.
And remove onednn concat limitation for depdendency input should be
onednn impl.

Signed-off-by: hyunback <hyunback.kim@intel.com>
2022-02-17 17:25:55 +09:00
Taylor Yeonbok Lee
d35335193a
[GPU] Adjust build batch size to 9 from 10 due to the compiler limitation w.r.t the entire module size (#10450) 2022-02-17 11:01:31 +03:00
Anastasia Kuporosova
861d43e06d
[Python API] Fix benchmark hanging (#10457) 2022-02-17 10:59:55 +03:00
Liubov Talamanova
be6a3c34f1
[POT] Throw exception for IRv10 (#10345)
* [POT] Throw exception for IRv10

* Update reference models

* Updated AC framework name from dldt to openvino
2022-02-17 10:54:08 +03:00
Vladimir Dudnik
29883a152a
fix 79520 (#10449) 2022-02-17 10:52:30 +03:00
Egor Shulman
ff293f5560
[CPU] Disable display of constant layers in PerfMap (#10307) 2022-02-17 10:51:07 +03:00
Egor Duplensky
541627d319
[CPU] [SANITIZER] Avoid possible stack-use-after-scope (#10377) 2022-02-17 10:27:58 +03:00
Ivan Tikhonov
3597ae61f9
Fix increased build time and memory consumption caused by multiple ov::Any instantiation (#10452)
* Fix increased build time and memory consumption caused by multiple instansion of ov::Any.

* delete unused method, correct exception message

* codestyle

* Resolve review comment

* fix exception: throw it in else branch
2022-02-17 09:08:55 +03:00
Gleb Kazantaev
926460e603
Fix Coverity issues (#10427) 2022-02-17 08:54:57 +03:00
Mateusz Tabaka
ab4a11b3bd
Remove unnecessary AutoBroadcastSpec parameter in MatMulMultiplyFusion (#10005) 2022-02-17 08:51:32 +03:00
Julia Kamelina
1fc61299c8
update omz submodule (#10441) 2022-02-17 00:50:21 +03:00
Tomasz Dołbniak
90a100d5f6
Default opset bump in ONNX FE (#10437) 2022-02-17 00:47:07 +03:00
Fedor Zharinov
00abcbacc4
Fix for Layout and image_info related issues (#10258)
* bugfix78627

* stylefix

* fix
2022-02-17 00:42:51 +03:00
Maxim Vafin
5cadee20eb
Fix issue with constants having inputs in TF FE (#10393) 2022-02-16 20:40:23 +03:00
Andrey Zaytsev
abeb910ce2
Removing the old Intel logo from docs (#10429)
* Added info on DockerHub CI Framework

* Feature/azaytsev/change layout (#3295)

* Changes according to feedback comments

* Replaced @ref's with html links

* Fixed links, added a title page for installing from repos and images, fixed formatting issues

* Added links

* minor fix

* Added DL Streamer to the list of components installed by default

* Link fixes

* Link fixes

* ovms doc fix (#2988)

* added OpenVINO Model Server

* ovms doc fixes

Co-authored-by: Trawinski, Dariusz <dariusz.trawinski@intel.com>

* Updated openvino_docs.xml

* Updated the link to software license agreements

* Revert "Updated the link to software license agreements"

This reverts commit 706dac500e.

* Removed the Intel logo

Co-authored-by: Trawinski, Dariusz <dariusz.trawinski@intel.com>
2022-02-16 17:26:26 +03:00
Yuan Xu
4f000b780d
update pypi installation (#10217)
* Add Overview page

* update pypi installation

* Revert "Add Overview page"

* integrate review comments

* update formatting

* Update docs/install_guides/installing-openvino-pip.md

Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>

* Update docs/install_guides/installing-openvino-pip.md

Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>

* Update docs/install_guides/installing-openvino-pip.md

Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>

Co-authored-by: Adrian Boguszewski <adekboguszewski@gmail.com>
Co-authored-by: Andrey Zaytsev <andrey.zaytsev@intel.com>
2022-02-16 17:09:56 +03:00
Egor Shulman
5bf9631073
Fixed ProfilingInfo layer status (#10342) 2022-02-16 16:10:19 +03:00
Anton Grishin
05650551b7
[GNA] Fix static analyzer issues (#10379)
* fix incorrect braces

* move pointer check

* add pointer check to VerifyConcat

* Prevent iterator invalidation
2022-02-16 15:46:32 +03:00