Commit Graph

4360 Commits

Author SHA1 Message Date
Maxim Andronov
2b871eb582
[CPU] prohibit fusing if dropped node contain > 1 child edges (#6705) 2021-07-30 11:47:47 +03:00
Aleksandr Voron
e3e2ee4d74
Update dependencies.cmake (#6861) 2021-07-30 11:27:16 +03:00
Anastasiya Koryachikhina
4d7f6c54ef
Change logic for using multiple models in cc tests (#6618)
* Change logic for using multiple models in cc tests

* removed extra key

* moved multiple infer to main function

* moved multiple infer to main function

* change test_config.yml and work with this file

* change test_config.yml and work with this file

* removed extra key

* changed separator symbol

* removed extra param from output

* Rollback comments in test_config.yml

* removed extra key in yml config file, change logic to save .npz for all model seperate instead of common npz file, change logic in parser arg

* removed extra key in yml config file, change logic to save .npz for all model seperate instead of common npz file, change logic in parser arg

* changed save path for infer result, deleted extra param

* updated test_config.yml

* rollback old way to get bin_path

* changed work with save path for inference result

* add empty line in the end for config file

* rollback line order

* removed extra param allow_pickle=True in loading .npz file

* removed mkdir in run_infer

* uncomment resnet model

* added empty line

* changed save folder for cc result
2021-07-30 10:30:38 +03:00
Ivan Tikhonov
bc06279825
Fix clone_function: clone ngraph::Variables (#6804)
* fix clone function

* ngraph codestyle

* fix copy function for assign/read value v3

* add unit test
2021-07-30 10:19:12 +03:00
Maxim Vafin
5920cf849d
Implement paddle frontend methods for partial conversion (#6784)
* Implement paddle frontend methods for partial conversion

* Apply feedback

* Fix codestyle

* Remove normalize implementation and change convert signature for partialModel case

* Apply review feedback

* Apply review feedback

* Fix onnx importer convert signature

* Fix codestyle

* Add test of unsupported op

* Fix code style

* Remove "normalize()" tests

* Fix onnx tests

* Fix build

* Fix merge conflicts
2021-07-30 10:14:18 +03:00
Patryk Elszkowski
0861a5c910
Validate speedup (#6779)
* Add minor speedup changes.

* inline clip

* reduce clip calls

* more Interval::size - move to header

* terminate instead of throwing exception

* back to throw exception when element type was not found

* rename variable
2021-07-30 08:59:36 +03:00
Yury Gaydaychuk
7ab92b5845
[CPU] Adaptive pooling implementation (#6594) 2021-07-29 20:41:24 +03:00
Anastasia Popova
e70e7e1e9d
Shell implementation for RandomUniform. (#6782)
* Added shell implementation for RandomUniform.

* Small correction.

* Small correction.

* Corrected wrong type.

* Corrected error message, corrected setters.
2021-07-29 18:19:54 +03:00
Gleb Kazantaev
a3d9f00d98
Enable LinOpFusion To MO Backend (#6838) 2021-07-29 18:16:49 +03:00
Gleb Kazantaev
0a1cad52ab
Avoid Constant data copy inside Reshape constant folding (#6410)
* Avoid Constant data copy inside Reshape constant folding

* Fix Codestyle

* Updated Squeeze, Unsqueeze cf

* Deprecate set_data_shape method

* Fix Pruning
2021-07-29 17:51:25 +03:00
Egor Shulman
ab8d046642
[CPU] NormalizeL2 - unsorted axes are not supported (#6602) 2021-07-29 17:30:38 +03:00
Xie Zhengtian
513a7504a6
[AUTO] Update Auto-device plugin selection logic (#6335)
* Update selecting device logic to support Myriad and VPUX for auto-device plugin

Signed-off-by: Zhengtian Xie <zhengtian.xie@intel.com>

* Update selecting logic in auto-device plugin

Signed-off-by: Zhengtian Xie <zhengtian.xie@intel.com>
2021-07-29 15:57:45 +03:00
Andrey Somsikov
d498653d1f
Fix clang sanitizer build (#6853) 2021-07-29 15:45:27 +03:00
Mikhail Nosov
d65dbd1200
Fix Windows_2019 CI (#6854) 2021-07-29 14:55:33 +03:00
Alexey Lebedev
4c0c184d3f
[IE PYTHON] Release GIL in some functions (#6236)
* Release GIL in load_network

* release gil in infer, wait and get_idle_request_id

* release gil in read_network and IECore.__cinit__

* release GIL in properties

* Release GIL in infer_async

* Add test

* Fix test

* Fix test
2021-07-29 14:28:11 +03:00
Ivan Tikhonov
6673a90659
MO DeformableConv: fix a set of attributes (#6806)
* MO DeformableConv: fix a set of attributes

* codestyle

* resolve review comments

* resolve review comments
2021-07-29 13:49:12 +03:00
Ilya Churaev
233bc540ef
Do not throw an exception if we have 2 result op for one tensor (#6844) 2021-07-29 13:16:20 +03:00
Mikhail Nosov
868fad33ab
[MO] Add support to moc_frontend of ":" as delimiter for --input and --output (#6543)
* [MO] Add support to moc_frontend of ":" as delimiter for --input

Additions:
Changed default logic for 'Place::get_in(out)put_port' to return nullptr
Changed default logic for 'InputModel::get_place_by_tensor(operation)_name' to return nullptr

* Corrected comments in code

* Missing empty line

* Clang format fixes

* Fix review comments

* Updated test to verify review comments fixes

* Update unit tests after rebase

* Apply review comments
2021-07-29 13:09:51 +03:00
Mikhail Nosov
a95d59014c
Rename "pdpd" frontend to "paddle" (#6842)
* Rename "pdpd" frontend to "paddle"
Also renamed mo_pdpd.py+main_pdpd.py to mo_paddle.py+main_paddle.py

Class names were not renamed

* Update package_BOM.txt
2021-07-29 11:43:07 +03:00
Patryk Elszkowski
0f6e4afa71
update FakeQuantizeLayerTest - add broadcast (#6813) 2021-07-29 07:46:19 +03:00
Alexandra Sidorova
709d7b8aab
[CPU] Added separate inference for pc, splitted nodes (#6511) 2021-07-28 22:42:51 +03:00
Ilya Churaev
62ed84e880
Avoid copy of data when create Constant from HostTensor (#3388)
* Avoid copy of data when create Constant from HostTensor

* Fixed build
2021-07-28 18:27:27 +03:00
Mikhail Nosov
5d3ab3108f
Paddlepaddle unit tests CI fixes (#6820)
Win+Lin: add "clean build dir" step to ensure generated test models are taken from install
Windows fix: execute "pip show" on same drive where python is installed
Windows: execute tests with FE_TEST_MODELS env variable
2021-07-28 17:34:30 +03:00
Alexandra Sidorova
1aa58b4c7d
[CPU] Enabled MatMul+Transpose transformations and reduced MatMul inference overheads (#6570) 2021-07-28 16:11:41 +03:00
Bartosz Lesniewski
1471095bdb
Revise Less (#6728)
* Revise spec

* Comparison backend test POC

* Split Comparison ops tests into separate files

* remove comparison.cpp, remove unused imports, replace for_each with range based for

* remove unnecessary ngraph:: prefixes

* Fix links in spec

* Add Less to trusted ops list

* Add missing ','

* Use builder in backend tests

* Remove old backend tests for less, equal
2021-07-28 15:36:07 +03:00
Ilya Sharikov
6ee9285fe5
Added Windows OS support to MemCheckTests (#6684)
* Added Windows OS support to MemCheckTests

* Move system headers to cpp file

* Moved includes for windows to cpp file. Was rewrited function to get memory info

* Was rewrite functions for getting information about memory

* Moved implementation run_in_process to cpp file

* Fixed implementation of run_in_process
2021-07-28 13:06:16 +03:00
Edward Shogulin
c0eb700024
[LPT] Exception fix (#6824) 2021-07-28 00:08:23 +03:00
Yury Gaydaychuk
1f24cc9cea
[CPU] nms disabled (#6822) 2021-07-27 21:16:13 +03:00
Andrey Somsikov
2e26dfd4dd
Fix sanitizer build (#6790)
Use default linker for sanitizer builds

Compiler out of memory tracking variables in GPU tests.
This trigger recompilation without tracking. Disabling
tracking globally as not required option.

UBSan build is failing on maybe uninitialized variable.
Temporarily disable the check.
2021-07-27 15:29:34 +03:00
Alexander Zhogov
16dc8d4c52
Azure CI: Add building Python API on Windows (#6799)
* Azure CI: Add building Python API on Windows

* Add Python requirements

* Fix Python ver

* Fix
2021-07-27 13:34:33 +03:00
Maksim Kutakov
e47a85b427
[CPU] Get/Set Blob overhead has been eliminated. (#6737) 2021-07-27 13:22:09 +03:00
Mateusz Bencer
dc5f44e929
Integration ONNX Editor with FE API (#6773) 2021-07-27 12:20:39 +02:00
Gleb Kazantaev
9acedbdacf
Enable TransposeSyncinc and BroadcastElementwiseFusion in MOC Backend (#6796) 2021-07-27 12:24:28 +03:00
Ilya Churaev
71d826e9fa
Fixed CI after #6199 (#6814) 2021-07-27 12:13:10 +03:00
Patryk Elszkowski
7bbf3775bd
initialize NodeBuilder registry by IIFE instead of double lock pattern (#6753) 2021-07-27 07:53:29 +03:00
Szymon Durawa
f5fe8550d9
Remove deprecated convolution API. (#6732)
* Remove deprectaed convolution API.

* Apply correct style.
2021-07-27 07:36:10 +03:00
Yury Gaydaychuk
e5a990e356
[CPU] RoiPooling handles border proposals correctly (#6199) 2021-07-26 23:35:27 +03:00
Ilya Lavrenov
477a1ae04f
Fixed compilation for Android w/o PDPD (#6801) 2021-07-26 20:25:39 +03:00
Yunji Kim
f5666fb3e1
[GPU] Add new operation GatherElements to IE clDNN plugin (#6676) 2021-07-26 17:52:27 +03:00
Dawid Kożykowski
b4ad7a1755
Refactor visitor tests for BinaryElementwiseArithmetic ops (#6667)
* add binary_elementwise file

* change binary_elementwise.hpp to binary_ops.hpp

* migrate mod operation test o typed template test

* add tests for remaining binary ops

* remove comment

* fix formatting to match clang-format

* add RVO-exploit string concatenating andbeautify the code

* add validation for attributes number

* add missing visit_attributes() calls

* add missing 4th param to NGRAPH_RTTI_DEFINITION calls

* fix formatting to match clang-format
2021-07-26 15:49:14 +03:00
Dawid Kożykowski
1a92a69515
Refactor visitor tests for UnaryElementwiseArithmetic ops (#6756)
* remove namespace dependencies from header file

* add RVO-exploit string concatenating

* update acosh test

* update erf test
2021-07-26 15:46:31 +03:00
Ilya Churaev
462c2f9945
Avoid replace parameter in reshape method (#6741)
* Avoid replace parameter in reshape method

* Added comment

* Use set_partial_shape()
2021-07-26 15:11:31 +03:00
Ilya Lavrenov
37c619b0b8
Fixed compilation with Intel compiler (#6788)
* Fixed compilation with Intel compiler

* Fixed template containers for VPU

* Fixe one-dnn

* Updated submodule
2021-07-26 12:42:05 +03:00
Gleb Kazantaev
83a3e083a1
Add more passes to the MOC backend (#6749) 2021-07-26 12:33:23 +03:00
Evgenya Stepyreva
b907c3b84f
set_output_type speedup (#6754)
* set_output_type speedup

* style

* Final optimization

* Removed extra include, removed unnecessary lock_guard

* Typo

* Apply suggestions from code review

Co-authored-by: Mikhail Nosov <mikhail.nosov@intel.com>

* Update ngraph/core/include/ngraph/descriptor/tensor.hpp

Co-authored-by: Mikhail Nosov <mikhail.nosov@intel.com>

Co-authored-by: Mikhail Nosov <mikhail.nosov@intel.com>
2021-07-26 12:00:13 +03:00
Andrey Somsikov
feb1eaef05
Fix file handle leak in cnpy (#6380)
* Fix file handle leak in cnpy


Co-authored-by: Roman Donchenko <roman.donchenko@intel.com>
2021-07-26 11:54:39 +03:00
iliya mironov
16d0c28314
Add check if fastjsonschema installed (#6775) 2021-07-26 11:39:44 +03:00
Mikhail Nosov
ea60879a8b
PDPD: Install generated PDPD test models as 'test' component (#6783)
* Install generated PDPD test models as 'test' component

Needed when test package is installed on another machine which don't have generated models

To search installed models, 'FE_TEST_MODELS' env variable is introduced

* Clang fixes

* Fix PDPD_Places test from latest master
2021-07-26 10:12:26 +03:00
Andrei Molotkov
9b706711fe
[GPU] Refactoring of pre_replace_deconv pass (#6218) 2021-07-26 10:02:21 +03:00
Ilya Sharikov
33dfcd62c2
Add --verbose option to infer_tool (#6770)
Detailed infer results printing overload output and freeze sea_runtool.
Hiding debug messages by default.
2021-07-26 09:00:56 +03:00