Commit Graph

3167 Commits

Author SHA1 Message Date
Vladislav Golubev
8b4837ea62
[LPT] Correct handling of Dq after operations with several outputs (#4797)
* [LPT] Dq after Split/VariadicSplit fix

* [LPT][TESTS] Transformations after split tests
2021-03-25 23:19:14 +03:00
Evgeny Lazarev
522ad39a48
Tf 2.0 od api models with loop (#4772)
* Do not run transformations for TF 2.X OD API models recursively (needed for models with Loop operation)

* Added anchor front transformation to group all TF OD API transformations. Added new necessary dependencies from KerasRNN transformations related to While support

* Added JSON configuration files for TF 2.4 OD API SSD and EfficientDet models

* Updated documentation with table of supported TF 2.x OD API models

* Improved visualization of the dependency graph

* Updated version of the pre-processing transformation for TF 2.4 OD API models

* Fixes in the TF 2.x OD API models conversion

* Fixed order of applying mean/scale values for TF 2.X OD API pre-processing

* Updates to the documentation

* Fixes for the preprocessor block transformation for the TF OD API models

* Added code comments

* Fixed bom file

* Unit tests for the TF 2.4 OD API ObjectDetectionAPIPreprocessor2Replacement transformation

* Code cleanup

* Updates to the documentation on how to convert TF OD API models and graph dumper

* Added assert to make sure that operations in the `get_specific_ops_with_const_inputs` has exactly 2 inputs
2021-03-25 17:56:35 +03:00
Edward Shogulin
26a4022672
[CPU] Erf JIT implementation (#3790) 2021-03-25 17:50:00 +03:00
Ilya Sharikov
7681f86f50
Fixed double slash in c_api inference engine installation (#4931) 2021-03-25 17:12:16 +03:00
Ilya Lavrenov
8e261de0a8
Added PreprocessConversionTest tests (#4946)
* Added PreprocessConversionTest tests

* Disabled tests on GPU: CVS-51764

* Disabled some tests on VPU and TEMPLATE

* Support for input layout conversions in TEMPLATE plugin
2021-03-25 16:31:33 +03:00
Maxim Vafin
2a52747b03
Fix reading FP16 scalars from TensorFlow (#4935) 2021-03-25 15:59:02 +03:00
Irina Efode
fa37277e3e
[IE TESTS] Allow to save Conformance report with unique names (#4844)
* [IE TESTS] Lock report file (between processes and threads)

* Fix condition

* t

* Fix CI

* Fix ci

* win

* Fix win

* Fix reading logic

* Fix rights

* d

* Fixes

* d

* ddd

* s

* dd

* Allow to write reports by unique name

* tp

* fix ci

* remove extra lines
2021-03-25 15:29:20 +03:00
Liubov Batanina
79905e9cf7
ROIAlign: added consistency of input data types (#4936) 2021-03-25 14:59:44 +03:00
Ilya Churaev
c657574a82
Optimize nGraph linker dependencies (#4929)
* Refactored nGraph tests dependencies

* Fixed CI

* Fixed ONNX dependencies

* Fixed comments
2021-03-25 13:30:18 +03:00
Ilya Lavrenov
f205ba9459
Fixed pyngraph compilation with python 3.9.2 (#4940) 2021-03-25 13:15:01 +03:00
Dmitry Kurtaev
64874fa69c
Remove __clang__ macros from headers (#4575)
* Fix headers for Clang compilation

* Parameter::Any destructor

* Force RTTI for any compiler

* Parameter::RealData RTTI

* Revert Parameter::Any
2021-03-25 12:26:43 +03:00
Sergey Lyubimtsev
222a709c6c
NGraph python api cannot be built w/o onnx importer (#4880)
* NGraph python api cannot be built w/o onnx importer

* add target condition
2021-03-25 12:26:35 +03:00
Maxim Vafin
d40a607ca0
Support Pad-2 in opset-11 ONNX model (#4886)
* Support Pad-2 in opset-11 ONNX model

* Add unit test for pad

* Apply review feedback
2021-03-25 12:05:28 +03:00
Ilya Lavrenov
d77869ba71
Properly apply patch to ONNX (#4939) 2021-03-25 11:54:48 +03:00
Vladimir Gavrilov
af95452026
nGraph shell for the operation DFT (#4625)
* Written the header file for the nGraph operation DFT.

* Small change.

* Started to write the implementation of the nGraph shell of the operation DFT.

* Started to write void op::v7::DFT::validate_and_infer_types().

* Small fixes.

* Code style fixes.

* Written the draft of the shape infer for the nGraph operation DFT.

* Small fixes.

* Code style fixes.

* Added DFT into opset7 table.

* Some additions.

* Small fixes.

* Code style fix.

* Some fixes.

* Some fix.

* Small fixes.

* Started to write shape infer tests for the nGraph operation DFT.

* Written shape infer tests for the nGraph operation DFT.

* Some code style fixes.

* Small fix.

* Code style fixes.

* Code style fixes.

* Deleted unused variables.

* Added support for negative axes.

* Started to write IE IR Reader tests for the nGraph operation DFT.

* Small fix.

* Added the second IE IR Reader test for the nGraph operation DFT.

* Small fix.

* Added the third IE IR Reader test for the nGraph operation DFT.

* Corrected Doxygen comment.

* Started to rewrite DFT type_prop tests as parametrized tests.

* Small fixes.

* Some additions.

* Small fix.

* Small fix.

* Some tests were written as parametrized tests. Some code style fixes.

* Code style fixes.

* Continued to rewrite tests for DFT as parametrized ones.

* Some deletions.

* Some additions.

* Deleted redundant tests.

* Started to rewrite some tests.

* Some changes.

* Deleted commented code.

* Started to convert tests for the case non-constant axes into parametrized tests.

* Rewritten tests for the case non-constant axes as parametrized tests.

* Started to convert tests for the case non-constant signal_size into parametrized tests.

* Rewritten tests for the case non-constant signal size as parametrized tests.

* Added checks for number of inputs.

* Small fixes.

* Small fixes.

* Refactored shape infer and corrected tests.

* Some refactoring.

* Now the function validate() is protected.

* Small refactoring.

* Fixed typo.

* Added some comments.

* Fixes in infer function.

* Added test.

* Fixed test.
2021-03-25 11:40:37 +03:00
Ilya Churaev
b76e965e95
Remove obsolete cmake script and test data (#4944)
* Remove obsolete cmake script

* Removed obsolete models

* Removed obsolete generators

* Remove obsolete markdowns
2021-03-25 07:45:47 +03:00
Alexey Suhov
a748c26fee
Align copyright notice in cpp and cmake source files (CVS-51320) (#4950) 2021-03-25 02:40:09 +03:00
Mateusz Tabaka
c9d0292929
Add DilatedConvolutionConverter, BatchToSpaceFusion and SpaceToBatchF… (#4689) 2021-03-24 16:00:30 +01:00
Ilya Churaev
f0e574903a
Int4 support for nGraph (#4712)
* Start int4 type

* Changed int4 type

* Fixed code style

* Added u4 type

* Added tests

* Add U4 I4 precisions to the IE

* Removed redundant tests

* Added convert precision

* Fixed tests

* Fixed C tests

* Fixed logic for signed values

* Fixed nGraph tests

* Added more tests

* Fixed code style

* Update inference-engine/src/transformations/src/transformations/convert_precision.cpp

Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>

Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
2021-03-24 16:17:43 +03:00
Piotr Szmelczynski
b62f49b66a
Revise power (#4901)
* add power type prop test file to CMakeList

* add unit tests

* create power type_prop tests

* Update copyright year

* fix copyright year
2021-03-24 15:29:16 +03:00
Bartek Szmelczynski
d5d0fca624
Revise minimum (#4784)
* add SLT for serialization for minimum/maximum ops, and add unit test for minimum op

* style fix

* change 2020 to 2021 intel

* fix copyright dates
2021-03-24 15:28:29 +03:00
Aleksandr Korolev
4cac858346
[IE][VPU][TESTS]: Fix for myriad stress test (#4902)
Reused inputs for cycling infer requests in MultipleAllocations test.
2021-03-24 14:25:06 +03:00
Krzysztof Bruniecki
985adb8223
Sync model loading from different threads using single Core (#4842)
- Enable tests
2021-03-24 13:44:23 +03:00
Mateusz Tabaka
e7b9b021ae
Implement ConvtoBinaryConv transformation (#4038) 2021-03-24 11:06:26 +01:00
Vladimir Gavrilov
b99345c320
nGraph shell for the operation IDFT (#4660)
* Written the header file for the nGraph operation IDFT.

* Written an implementation of the nGraph shell of the operation IDFT.

* Added IDFT into opset7.

* Some additions.

* Written shape infer tests for the nGraph operation IDFT.

* Added support for negative axes.

* Written IE IR Reader tests for the nGraph operation IDFT.

* Code style fixes.

* Small fix.

* Fixed Doxygen comment.

* IDFT type_prop tests were rewritten as parametrized tests.

* Added checks for number of inputs.

* Deleted unsupported types for 'axes' input.

* Checks for inputs were moved into the separate function.

* Some fixes.

* Some fixes in tests.

* Some fixes in tests.

* Some refactoring.

* Some fixes.

* Some fixes in shape infer function.

* Fixes in shape infer and corresponding tests.

* Small fix.
2021-03-24 13:00:59 +03:00
Andrew Bakalin
66cca724f6
[VPU] Add option to disable preprocessing check inside the model (#4616) 2021-03-24 12:57:02 +03:00
luocheng
29873c0a1d
wait cpu executor destroy (#4796) 2021-03-24 12:55:45 +03:00
Mateusz Tabaka
9b829d2884
Introduce AddFakeQuantizeFusion and MulFakeQuantizeFusion transformat… (#4102) 2021-03-24 09:19:37 +01:00
Anastasia Popova
83e5bde4ea
Shell implementation for Roll operation. (#4843)
* Added shell implementation for Roll operation.

* Added test, scalar check corrected.

* Code style correction.

* Comment fixed.

* Removed redundant virtual.

* Added parentheses.

* Fixed tests, added axes values check.

* Fixed scalar check, added axes check.

* Added comment.

* Added static shape checks, added more tests.

* Corrected error messages.

* Corrected error messages.
2021-03-24 09:58:13 +03:00
Ilya Sharikov
568f320cbc
Fixed installation ngraph core (#4920) 2021-03-24 06:38:12 +03:00
Ilya Lavrenov
89203c00b0
Removed obsolete NGRAPH_COMPONENT_PREFIX (#4921) 2021-03-24 06:37:11 +03:00
Liubov Batanina
a09218d2d7
Added float16 to gather_tree reference (#4917) 2021-03-24 06:14:54 +03:00
Alexey Suhov
345f816709
[README.md] change latest release to 2021.3 2021-03-23 22:24:38 +03:00
Maxim Vafin
b2f3243387
Add MVN fusion (#4647)
* Add MVN fusion

* Fix map for Or pattern

* Use count instead of find in MVN fusion

* Apply review feedback

* Apply review feedback

* Fuse patterns

* Fix Win build

* Apply feedback
2021-03-23 19:27:53 +03:00
Anton Pankratv
3b3d9a0989
Used IE_THROW macro (#4869)
* Added ie throw macro

* Used IE_THROW macro
2021-03-23 18:57:12 +03:00
Mikhail Ryzhov
c3fc40052c
Python 3.8 support (#4629)
* Python 3.8 support

* Get path from setupvars
2021-03-23 17:32:08 +03:00
Anna Khakimova
87000ed1ca
Pre-processing: SIMD optimization of the Resize 8UC1 (#4804)
* * Pre-processing: SIMD optimization of the Resize 8UC1

* * Refactored and added new universal intrinsics.

* fix for ARM32 issue

* * Refactoring.
2021-03-23 17:29:23 +03:00
Mikhail Nosov
5caa706334
Enable cache support for Myriad plugin (#4868)
* Enabling cache support for Myriad plugin

Remaining issues:
1) 49309 - support DEVICE_ARCHITECTURE
2) Fix 51472 for import/export issue with batch size > 1

* Added 'options' to DeviceArchitecture and more clarifications
2021-03-23 17:28:52 +03:00
Mikhail Nosov
86eebbdfac
Python API for LoadNetwork by model file name (#4896) 2021-03-23 16:58:20 +03:00
Alina Alborova
1d76ab7f55
Review Inference Engine Extensibility docs (#4267) 2021-03-23 15:28:16 +03:00
Ilya Churaev
3ee41375a3
Clang-format check for itt collector (#4893)
* Enable clang_format for ITT collector

* Fixed code style

* Bad code style

* Enable clang-format for ITT collector

* Revert "Bad code style"

This reverts commit 82cbd40539.

* Fixed build
2021-03-23 15:08:51 +03:00
Szymon Irzabek
3672fb9817
Enable CNN2D tests for GNA library 2.1 (#3917)
* [GNA] Enable CNN2D tests for GNA library 2.1

* [GNA] Fix test skipping decision
2021-03-23 14:52:45 +03:00
Patryk Elszkowski
a1ed4f1b49
Update Pad specification (#4799)
* udpate Pad specyfication

* apply PR suggestions

Co-authored-by: Patryk Elszkowski <patryk.elszkowki@intel.com>
2021-03-23 13:12:58 +03:00
Ivan Tikhonov
e877272eca
Clarification of Low Latency Transformation and State API documentation (#4771)
* Assign/ReadValue, LowLatency and StateAPI clarifications

* Apply suggestions from code review: spelling mistakes

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

* fixed wording

* Resolve review remarks

Co-authored-by: Anastasiya Ageeva <anastasiya.ageeva@intel.com>
2021-03-23 12:40:42 +03:00
Bartek Szmelczynski
910e83b2a2
Rewrite implementation for input and output description (#4506)
* initial split of on_adapter function in order to check if all checks will pass

* remove blank line and change if statements for on_apdapter for void type

* add type checking instead of name checking

* add bracket two if/else if statements and move helper functions to private

* fix style

* revert removing ngarph check for empty pramas and result

* move map_type_from_body method to private, add const& function arguments
2021-03-23 06:16:20 +03:00
Evgeny Lazarev
062bfd88d3
Fix Gelu parameter name (#4895)
* Rename attribute of the "Gelu" operation from "approximation" to "approximation_mode"

* Updated MO file for coverage check
2021-03-22 19:35:32 +03:00
Andrey Dmitriev
9d69c0e0ec
[GNA] Multithread speech sample with iname (#4774) 2021-03-22 17:14:15 +03:00
Taylor Yeonbok Lee
50e1f43e26
[IE CLDNN] Fix permute reorder fusing error (#4876) 2021-03-22 16:33:46 +03:00
Katarzyna Mitrus
b8f36ec354
MatMul reference implementation refactoring (#4671)
* MatMul backend tests

* Single layer tests

* Style apply

* Updaye IE_CPU manifest

* New dot implementation

* Use new dot in MatMul reference

* Fix output batch offset

* Style apply

* Relax tests tolerance

* Remove legacy dot reference file

* Remove usage of broadcast builder

* Add one more broadcast test

* Remove NGRAPH_SUPPRESS_DEPRECATED

* Style apply

* Few more MatMul single layer tests

* Update IE tests manifest

* Move variable declarations to inner loops

* Add const to variables

* Apply review suggestions

* Reuse vector for transposed and broadcasted data
2021-03-22 15:27:31 +03:00
iliya mironov
e64d84b47b
Enable gpu tests. Disable fp16 cpu test. (#4856) 2021-03-22 14:47:07 +03:00