Commit Graph

6332 Commits

Author SHA1 Message Date
Mateusz Bencer
5fe228bc14
Intorduce json model analysis feature on the new path (#9618) 2022-01-18 11:18:19 +01:00
Ilya Churaev
a8381f71f9
Fixed coverity issue for inference component (#9688)
* Fixed coverity issue for inference component

* Removed noexcept
2022-01-18 12:43:10 +03:00
Gleb Kazantaev
18eaaedb39
Fix If operation validation function (#9726) 2022-01-18 12:41:50 +03:00
Egor Duplensky
3a9dba7362
[CPU] [DEBUG CAPS] Implicitly enable perf counters for verbose mode (#8754)
Also refactor debug capabilities configuration.
Now debug caps config parameters are the part of plugin configuration.
2022-01-18 12:32:53 +03:00
Svetlana Dolinina
71b7c7000a
Changed permutes mechanics for Kaldi (#5744)
* changed permutes

* fixed permutes

* fixed kernel

* fix transpose after convolution

* fix for convnet

* insert transposes for all convolutions and poolings

* refactor transformations;
added unit tests;
removed old transformations for addinf permutes/reshapes

* fixed constant types

* fixes after merge

* fixed bug for rm_cnn4a: added correct time_dim for the first convolution

* added fix for timeheightconvolution: in this case we have correct time set in convolution kernel already

* minor review fixes: renamed transformation and file

* rename in test

* rename in test

* sort imports + couplt changes in comments

* review fixes: refactoring

* replaced recursive implementation by nx.topological_sort;
fixed comments

* minor fixes: comment + preserving node names
2022-01-18 12:10:51 +03:00
Anton Chetverikov
919ada63ad
Update undefined dimensions parcing im parameter extender (#9699) 2022-01-18 12:02:30 +03:00
Katarzyna Mitrus
18309852ba
Update Model::reshape to rethrow currently handled exception (#9723)
* Update Model::reshape to rethrow the currently handled exception

* Set expected type of exception in TestInvalidReshape
2022-01-18 11:49:02 +03:00
Tingqian Li
89f4a569d5
[CPU] Reuse the new nGraph shape infer functionality to the MKLDNN plugin (#9220) 2022-01-18 11:46:35 +03:00
Tingqian Li
822cb48a1e
[CPU] Add cache support to transpose node (#9513)
* Add cache support to transpose

* Re-enable test case

* comments apply
2022-01-18 11:29:32 +03:00
Ivan Vikhrev
a2cf98bebb
[IE Samples] json configuration reader and dumper for benchmark_app (#9648)
* added load_config and dump_config functions implemented with json library

* add warning, upd readme

* Update samples/cpp/benchmark_app/README.md

Co-authored-by: Fedor Zharinov <fedor.zharinov@intel.com>

Co-authored-by: Fedor Zharinov <fedor.zharinov@intel.com>
2022-01-18 11:22:47 +03:00
Nadezhda Ageeva
eb9e2c986a
GNA: move transformation test (#9716) 2022-01-18 11:17:21 +03:00
Yegor Kruglov
9b129b7c1a
[MO] Support MXNet Operations: batch_dot, LayerNorm, div_sqrt_dim (#7641)
* Add new operation support

* Update range_like replacer

* Move layer normalizer to middle

* Update bom file

* Update bom

* removed ArangeLike op

* updated bom

* added tests

* Updated docs

* comments relolving

* resolve documentation merge conflict

* arange_like op

* Revert "arange_like op"

This reverts commit a30f5bbb48.

* fixes in div_sqrt_dim

* comments resolving

* updated tests

* added batch_dot and layer_norm descriptions

* updated batch_dot comment

* updated comment

* move extractors to mxnet folder

* added replacer for batch_dot

* Revert "added replacer for batch_dot"

This reverts commit 8c0e52f7dc.

* return fully connected normalization

* fix typo

* updated fully connected normalization for mxnet

* changed assert message

* fixed gamma and beta shape incompatibility problem

* fixed imports, updated unittest

* resolve comments

Co-authored-by: iimironov <iliya.mironov@intel.com>
2022-01-18 11:12:30 +03:00
Tingqian Li
ff2df42339
[CPU] Add cache support for normalize (#9508) 2022-01-18 10:06:52 +03:00
Vladimir Paramuzov
4c36f8d16d
[GPU] Allow 3d fsv16 layout for quantize and reduce prims (#9697) 2022-01-18 09:44:16 +03:00
Maxim Andronov
dd8a073aa4
[CPU] FullyConnected leftovers (#9621) 2022-01-18 09:14:29 +03:00
Mikhail Nosov
26f222dea5
[Tests] Graph comparator - take sinks into accounts (#9695)
* Graph comparator - take sinks into accounts

Previously graph has been traversed only from Results, so any differences in 'Sinks' were not detected

* Fix functional tests

* Update after internal discussion

* Fix low_latency_test (addition to low_latency_v2_test)

* Fix typo
2022-01-18 09:01:05 +03:00
Gleb Kazantaev
72942471de
Added MO root directory search to simplify MO execution process (#9681)
* Added MO root directory search to simplify MO execution process

* Update sys.path

* Simplify implementation
2022-01-18 01:58:22 +03:00
Sergey Lyubimtsev
5ebbad9bcf
Fixes for scripts (#9640)
* fixes for scripts

* reduce to warning python bitness check
2022-01-17 21:54:05 +03:00
Nikolay Tyukaev
a6c97000c5
update requirements (#9714)
* update requirements

* fixes
2022-01-17 17:38:15 +03:00
Ilya Churaev
6c0cddec78
Removed redundant sources from core component (#9689)
* Removed redundant sources from core component

* Fixed CC build

* Removed redundant includes
2022-01-17 16:16:03 +03:00
Maxim Shevtsov
7ca9641ead
Auto batch ov2.0 API support+tests (#9672)
* additional OV 2.0 tests for Auto-Batching

* adding 2.0 way of creating the requests

* performance counters support
2022-01-17 16:00:35 +03:00
Yuan Hu
580a0f30eb
[AUTOPLUGIN] Add format string checker for log information (#9592)
* add limit format on snprintf

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* add limit on format

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* add test case

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* fix a bug for LOG_TRACE

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* remove debug info
2022-01-17 14:50:01 +03:00
Anton Chetverikov
394e5c5515
Change nGraph serializer behavior for dynamic dimensions (#8526)
* Change dynamic_to_static to dinamic_to_undefined function to use -1 values instead of upperbounds in serialized IRs

* Resolve conflict after src files transition

* Remove resolve_dynamic_shapes function

* Fix typo

* Fix codestyle
2022-01-17 14:43:09 +03:00
Nadezhda Ageeva
56581dbe2e
[GNA] Support new kaldi irs (#9474)
* Support new kaldi IRs (generated in NHWC layout)

* Update tests with activation and fq

* Cleanup

* Fix reordering FQ and MaxPool and problem with overflow

* Fix win

* Update src/plugins/intel_gna/transformations/unfuse_reshape_and_transpose.hpp

Co-authored-by: Elizaveta Lobanova <elizaveta.lobanova@intel.com>

* Update src/plugins/intel_gna/transformations/unfuse_reshape_and_transpose.cpp

Co-authored-by: Elizaveta Lobanova <elizaveta.lobanova@intel.com>

* Update inference-engine/tests/unit/gna/ngraph/transformations/gna_unfuse_reshape_and_transpose.cpp

Co-authored-by: Elizaveta Lobanova <elizaveta.lobanova@intel.com>

* Code review

Co-authored-by: Elizaveta Lobanova <elizaveta.lobanova@intel.com>
2022-01-17 14:16:23 +03:00
Andrey Somsikov
3c7589184d
Do not force gold linker for sanitizer builds (#9680)
Gold linker fails to link tests built with sanitizer
2022-01-17 13:14:42 +03:00
Maksim Shabunin
76a74f3f2d
Fixed compilation of user code with VS2015 (#9644) 2022-01-17 12:53:28 +03:00
Eugeny Volosenkov
0948a54f53
yolov4 conversion documentation is incorrect (#9143)
* Rewrite doc

* added addition info abot weights and cfg files

* fix path to cfg and weights

* fix description

* fix description

* fix description

* fix description

* fix description

* fix description
2022-01-17 12:32:18 +03:00
Luwei Zhou
a76483ddb2
[CPU] Enable CPU plugin cache for ShuffleChannels and SpaceToDepth (#9556)
* Space to depth.

* Shuffle Channels RT cache.

* Update

* Apllied reivew comments.

* Applied review comments.
2022-01-17 12:28:30 +03:00
Svetlana Dolinina
0ed3da0e86
add warning about order if both mean and scale set (#9284)
* add warning about order if both mean and scale set

* Update tools/mo/openvino/tools/mo/main.py

Co-authored-by: Anastasia Popova <anastasia.popova@intel.com>

* removed warning, added phrase in documentation

* fixed merge

* added phrase about order of ,mean and scale in MO help

* duplicate MO help phrase in doc

* Update docs/MO_DG/prepare_model/convert_model/Converting_Model.md

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>

* Update docs/MO_DG/prepare_model/convert_model/Converting_Model.md

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>

* Update docs/MO_DG/prepare_model/convert_model/Converting_Model.md

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>

* Update tools/mo/openvino/tools/mo/utils/cli_parser.py

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>

* Update tools/mo/openvino/tools/mo/utils/cli_parser.py

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>

* remove tabs

* fix in order of reverse, mean, scale

Co-authored-by: Anastasia Popova <anastasia.popova@intel.com>
Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>
2022-01-17 12:12:06 +03:00
Mikhail Letavin
f2be2c915f
[GPU] Implement dynamic shape case support for one dimension only (batch) via legacy dynamic batch functionality (#9314) 2022-01-17 12:02:52 +03:00
Sergey Lyubimtsev
98cbaf0f08
align numpy version range for openvino wheel package (#9669) 2022-01-17 12:00:04 +03:00
Anastasia Kuporosova
8f12b72f46
[Python API] Rename offline_transformations in the new api (#9634)
* [Python API] Rename offline_transformations in the new api

* remove comments

* one more place to rename

* fix mo building

* fix after merge

* fix for pot import
2022-01-17 11:57:00 +03:00
Krzysztof Bruniecki
97df59a4ab
[GNA] Restore deprecated options for GNA Plugin (#9611)
* Restore deprecatet options for GNA Plugin

* Apply review use INFERENCE_ENGINE_DEPRECATED macro

* Deprecate PWL_UNIFORM_DESIGN and PWL_MAX_ERROR_PERCENT

* Add doxygen deprecation message

* Use string for deprecated config keys to avoid compile time errors

* Use IE_SUPPRESS_DEPRECATED_START

* Fixup lint

* Fix indentation

* Usue future release instead 2022.2 in deprecated message, supress deprecation error in tests

* Fix test
2022-01-17 10:32:02 +03:00
Vladimir Paramuzov
1fc013208d
[GPU] Integer floor and floor_mod support (#9629) 2022-01-17 10:14:51 +03:00
Ilya Churaev
3d4e82dcd1
Added new tests for tensors element types (#9667) 2022-01-14 22:59:08 +03:00
Andrey Noskov
5ad242ecce
[GNA] Added reset before and after export (#8900)
* [GNA] Added reset before and after export

* [GNA] Added query states check
2022-01-14 19:06:44 +03:00
Daniil Lyakhov
4693f7b854
[Offline Transformations] EfficientNet_b0 Pruning Transformation Enabling (#8926)
* Add include guard to file_utils.cpp

* Rebase src

* Rename acc tests, fix rebase

* Revert debug changes

* Fix linter

* Move ac tests to new template

* Test updated

* Fix result operation are sharing output tensor with previous op

* Pruning test visualzation option

* Add ac support for all test cases with pruned kernels

* Remove redundant files

* Enable extended pruning logging by env variable

* Adjust pruning tests

* Remove pruning extended debug env var

* Enable init masks only in debug mode

* Set result mask to input tensor instead of output tensor by separate key

* Bug fix / Test coverage

* Fix comments
2022-01-14 18:43:38 +03:00
Vladimir Paramuzov
4734b17e52
[GPU] Fixed USM shared blob allocation type (#9662) 2022-01-14 14:35:08 +00:00
Daria Mityagina
c1e2e0344b
[ICV] Add flag to prevent IE from booting MX VPU (#9447)
An additional option whose purpose is to prevent the booting of the MX VPU when we want to generate some blobfile.
2022-01-14 17:15:51 +03:00
Irina Efode
720f78ebde
[IE TESTS] Skip tests with Default device (#9643)
* [IE TESTS] Skip tests with Default device

* test
2022-01-14 16:56:53 +03:00
Jan Iwaszkiewicz
e1696deb3a
[PYTHON] Expose If, Loop and TensorIterator API (#9603) 2022-01-14 14:30:30 +01:00
Roman Lyamin
baa1a8e59d
[GPU] Add batching buffer to new API (#9565) 2022-01-14 16:02:19 +03:00
Wang, Yang
c7d216db13
passing by reference to fix performance inefficiencies. (#9590)
* passing by reference to fix performance inefficiencies.

Signed-off-by: Wang, Yang <yang4.wang@intel.com>

* Passing by const reference.

Signed-off-by: Wang, Yang <yang4.wang@intel.com>
2022-01-14 15:33:58 +03:00
Jade Cho
9e01cbd532
[GPU] fix a minor bug of dump graph (#9659) 2022-01-14 15:06:58 +03:00
Tomasz Dołbniak
8b1514d49e
Set additional tensor names only if necessary (#9652) 2022-01-14 12:44:50 +01:00
Anton Dudchenko
eb299f819e
[IE][VPU] Rename and move VPU plugin to src/plugins (#9110)
* [IE][VPU] Rename and move VPU plugin to src/plugins

* Update some dependencies

* Change .md

* Update some dependencies

* Update some dependencies (again)

* I'll rename this one

* Remove ENABLE_MYRIAD arg

* Update XLink.cmake

* Update thirdparty CMakeList

* Minor fixes
2022-01-14 14:02:08 +03:00
Edward Shogulin
bd97d1edc6
[LPT] SharedValueAttribute extending (#9534)
* [LPT] SharedValueAttribute extending

* [LPT] tests + comments
2022-01-14 13:33:07 +03:00
Sergey Lyubimtsev
0ff88458f9
fix PyPI upload error: Invalid value for classifiers. Error: Classifier 'OSI Approved :: Apache Software License' is not a valid classifier. (#9656) 2022-01-14 13:09:12 +03:00
Anton Chetverikov
f4b0d40cd0
[MO] Skip rt_info in old graph comparator (#9638)
* Skip rt_info while graph comparing

* Remove incorrect comparing for rt_info
2022-01-14 12:49:12 +03:00
Anton Pankratov
790f02c0b1
Hetero creates single result node (#9572) 2022-01-14 12:26:39 +03:00