* Add new compile model api to support hash model memory (#14543)
* Add new compile_model api for ONNX RUNTIME OV EP
Allow compile_model() accept model/weight data.
* Update minor place
* Cache model if possible
* Compute hash based on model_xml and model_weight
* Update typo
* Change hash key computation for model's weights
* Resolve test case issue
* Use tensor replace blob for hash computation
* Fix hash computation isssue and add more test cases
* Fix a build issue caused by data format
* Add ov::loaded_from_cache checking for CompileModelLoadFromMemoryTest (#15030)
* Add ov::loaded_from_cache checking for CompileModelLoadFromMemoryTestBase
* Skip gna in skip_tests_config
* Ignore empty tensor for hash calculation (#15282)
* Ignore empty tensor for hash calculation
* Added test
* Fix conflict
* Trigger ci run test for customer_A branch
---------
Co-authored-by: River Li <river.li@intel.com>
Co-authored-by: Ilya Churaev <ilya.churaev@intel.com>
* Add test to verify add_extension with relative path (#15212)
* Add test to verify add_extension with relative path
* Fix code style
* Use std::string::find instead of std::regex
* Remove unnecessary include
* Add comments about generating relative path
* Don't add empty tokens when splitting path
* Resolve relative path to the extension
* Method available independent of build_type
* LessEqual Not Supported
* port #13997
Update support "Sign" layer to not supported.
Update support "Floor" layer to limited support.
Tickets: 88871 and 88693
* port #13995
Add "Floor" in Known Layers Limitations section.
Add "ConvTranspose" in Known Layers Limitations section.
Add "GatherElements" in Known Layers Limitations section.
Tickets: 88871 and 75208 and 88690
This patch fixes two issues:
* Typo in clang pragma directive
* Reverse the order of the arguments in OPENVINO_ASSERT() in
constant.hpp so this assert works for ov::bfloat16 and
ov::float16 types which have impelemented binary comparison
operators as the memeber functions
fixes the FP32 and FP16 swap of results.
* SPR-6448Y updated results
Updated all files with data for SPR 6448Y. New results in benchmark-data.csv, benchmark-performance-data.xls, benchmark-info-detailed.xls and platform-list_22.3.pdf
authored-by: Michael Frank Hansen <michael.f.hansen@intel.com>
* fix strided_slice_optimize for 5d new axis output, typo in strided_slice cl kernal, avoid handle_reshape when user is only one
* fix unit test of resample_eltwise_fusing_through/12
* add time_to_first_inference_result KPI as sum of inf latency values
* append time unit format after the value
* Revert "append time unit format after the value"
This reverts commit 4bdd82e06b.
* remove metric, to be consistent with rest ones
* added FEIL SCOPED_TIMER (It will report the sum of all stages)
* revert add time_to_first_inference_result KPI as sum of inf latency values
* place SCOPED_REGION(first_inference_latency) inside a block { } - tested and worked
* enable first_inference_latency for timetest_infer_API_2.0
tested and worked
* remove trailing whitespace
Co-authored-by: Daria Ilina <daria.krupnova@intel.com>
* [PyOV] Rewriting properties tests as hardware agnostic (#14684)
* [PyOV] Mark replacement of topk in test (#15095)
* undo unwished changes
Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
* added missed layout to string conversion
* added an unit test for layout_from_string
* added missing spaces around =
* replaced 95 with SCALAR
* moved layout_to_string into a new file
* changed to parametrized test
* add placeholder text
* try sphinx format for headers
* update resources and learn OV
* try ref linking
* try ref linking 2
* try ref linking 3
* change links to doc
* fix new line
* move TF frontend to "deployment" TOC st. 1
* move TF frontend to "deployment" TOC st. 2
* fix links for deployment section homepage
* add disclaimer on ARM support
* trying different links to api
* change API link
* Trying to hide an article in TOC
* change top arcticles content
* remove directive for tf article
* fix build error
Co-authored-by: Karol Blaszczak <karol.blaszczak@intel.com>
Shift to separating the Workflow section, including moving Run and Optimize inference inside Deploy locally
change several article and menu titles
minor additional restructuring
minor content tweaks
remove optimization introduction (may be brought back in parts later)
* several link fixes
* additional link fixes
* [TF FE] Add info messages in Model Optimizer and the preliminary user documentation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add test model with future operation
* Update docs/MO_DG/prepare_model/TensorFlow_Frontend.md
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* Apply code-review feedback: Recover use of dict
* Fix the rest of feedback
* Remove User documentation for TF FE
* Correct tests to provide required arguments
* Handle argv without framework argument in deduce_framework function
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* [ONNX FE] Support value freezing without specifying type
Implement gete_element_type method for ONNX InputModel.
Implement tests with freezing integer input and float models
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Update src/frontends/onnx/frontend/src/input_model.cpp
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
* Update src/frontends/onnx/frontend/src/input_model.cpp
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
* Update src/frontends/onnx/frontend/src/input_model.cpp
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
* Return only element type for model inputs and add ONNX FE tests
* Rename variable in the test
* Add syntax mark
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
* updated to get ockable memory for device inputs
* gpuFuncTest to set f64 scalar tensor
* aligned type names
* updated to get lockable memory for some specific precisions
* install_openvino_dependencies: Added ocl-icd for ubi8 image
* install_openvino_dependencies.sh: Added json-devel to pkgs_dev for ubi8
* install_openvino_dependencies: Added pkgconf-pkg-config to pkgs_dev
* install_openvino_dependencies: Moved pkgconf-pkg-config to high-level pkgs_dev
* install_openvino_dependencies: Fixed with case for centos7
* Update scripts/install_dependencies/install_openvino_dependencies.sh
applied comments
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
* [TF FE] Break the cycle in the different way
Earlier solution was incorrect due to inproper handling of forward edges cases
(edges going from parent to grand-child) for which topological sorting
of nodes can be interrupted.
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Break the cycle by NextIteration inputs
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Initial TOC config
TOC depth
prev-next btn
TOC initial links
* extensibility section
header changed to section
section populated
* media processing and CV lib
header changed to section
no content in the node article... needs to be added later
* openvino security
header turned into section
intro article used as the section parent
* ecosystem
header turned to section
section populated
* menu padding adjustment
* benchmarks and release notes
benchmarks moved to resources
release notes moved to top level
case studies link
* OpenVINO IR
IR articles put in a section
the parent article will require more content
* [TF FE] Support dynamic shape Placeholder freezing and PlaceholderWithDefault
Also, this PR contains reorganization of python unit tests for TF FE that
covers conversion and inference of different models in pbtxt.
This mini-infrastructure will be used in the future for TF FE support.
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Revert debug info
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Merging with a slight rewording of the last suggestion by Andrei. We still want to retain consistency between paragraphs in structure and naming. If any further polishing needs to be done, we can create a new PR, while it can be published ASAP.
* sync benchmark Python version with benchmark C++ version.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
* Update tools/benchmark_tool/openvino/tools/benchmark/main.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/utils/utils.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Enable dump/load to support ov::device::properties parsing in benchmark python version.
* Enable benchmark python version to parse configuration file with device properties setting for AUTO/MULTI.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update usage of benchmark dumping/loading configuration in README and help message.
* Update comment.
* Update comments of the code.
* Update format of '-load_config' help message.
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* change AUTO default hint to Latency
* Change the comment tput to latency according to wangyang opinion
* fix testcase for MULTI, the hint default value returned by MULTI is throughput
* Remove the redundant testcase and modify the name of the testcase that returns the default value of hint
* Code optimization according to bell opinion, add comments to testcase
* Correct the comments of testcase
* When user sets num_streams, AUTO/MULTI does not set the default hint to HW plugin
* Fix the problem that smoke_AUTO_MULTI_ReturnDefaultHintTest fails to run
* add num_streams and default hint mock testcase
* add auto default perf hint mock testcase
* docs: revise example and update outputs
* docs: revise example and update outputs
* docs: typo fix
* docs: rectify changes
* docs: add <a> tag
* Update README.md
* docs: change example to use asl-recognition model
* docs: fix <a> tag
* docs: update sample outputs
* docs: change example to use asl-recognition model
* docs: revise example to use asl-recognition model
Co-authored-by: Yuan Xu <yuan1.xu@intel.com>
* [GPU] Fix dead lock from exceeded max size of host memory allocation.
Dead lock hang issue because remote_context ignore the exception and
return null. Use lock_guard.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* [TF FE] Use output_shapes attribute for Placeholder
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add test model
* Apply code-review feedback and remove compile warnings
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Use prebuilt cmake in install-dependencies steps
* Removed building of custom python to save Azure CI time
* Adjusted conditions in cmake
* Download VPU custom kernels compiler only for x86
* Removed -pyver argument
* Improvements
* Detect glibc version
* Usage of new glic variable
* Fixed typo
* Increased glibc version for MYRIAD case
* Made OpenVINO cmake Linux flavors agnostic
* Detect glibc version on RedHat
* Use clcompiler only since ubuntu 18.04
* Universal way to detect glibc version
* Revert some changes
* Added support of AmaZon Linux 2
* Amazon Linux 2 support
Removes nav items from top of page and moves to left nav.
Uses a redirect to home.html to replace index.html.
Restores path variables
Removes debug print statement
Fixes path.
Signed-off-by: Kevin Putnam <kevin.putnam@intel.com>
* Use prebuilt cmake in install-dependencies steps
* Removed building of custom python to save Azure CI time
* Adjusted conditions in cmake
* Download VPU custom kernels compiler only for x86
* Removed -pyver argument
* Improvements
* Detect glibc version
* Usage of new glic variable
* Fixed typo
* Increased glibc version for MYRIAD case
* Made OpenVINO cmake Linux flavors agnostic
* Detect glibc version on RedHat
* Use clcompiler only since ubuntu 18.04
* Universal way to detect glibc version
* Revert some changes
* Apply suggestions from code review
fixes issue with GNA
* Apply suggestions from code review
fixed typo
Becuase pooling to use ngraph shape infer, extend 1D to 2D logic moveed ngraph opt to cldnn pooling.
This change has side-effect in onednn pooling, onednn pooling also doesn't support 1D.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* temp resolution to support model path for CPU in auto
Signed-off-by: fishbell <bell.song@intel.com>
* disable batch when load through model path
Signed-off-by: fishbell <bell.song@intel.com>
* add mark for future release
Signed-off-by: fishbell <bell.song@intel.com>
* implement step1: donotparse batch config if user not set explictly
Signed-off-by: fishbell <bell.song@intel.com>
* correct typo in case
Signed-off-by: fishbell <bell.song@intel.com>
Signed-off-by: fishbell <bell.song@intel.com>
* fix AddressSanitizer: stack-use-after-scope in ov::frontend::paddle::NodeContext::get_attribute_as_any
* fix cyclic reference of op place and decoder.
* fix code style
* fix code style.
* Updated install_dependencies.sh
* Added Fedora support
* Improvements for RPM generation
* Refactored work with gflags
* Apply suggestions from code review
* Improvements
* Several improvements for Ubuntu 18.04
* Improvements in gflags
* Update thirdparty/CMakeLists.txt
temporary install samples dependencies
* Fixed several mistakes
* Fixed issues with gflags
* Don't install dependencies on Linux
* Added nlohmann findinds on debian9 case
* Added FATAL_ERROR is gflags is not found in samples
* Fixes for samples on CentOS 7
* Cross-compiled debian packakges with proper libraries location
* Fixed include for old nlohman-json versions
* Fixed compilation with old json
* Applied review comments
* Added support of old nlohmann versions via submodule
* Temporary WA for CentOS 7
* Fixed compilation with old gflags
* Fixed compilation of C samples
* Fixed gflags on RHEL8.2 / Debian 9
* Use openvino pass graph_rewrite
* Use openvino pass pattern matcher
* Remove ngraph opsets
* Remove ngraph.hpp
* Remove ngraph includes
* Remove ngraph includes
* Use transformations API
* Remove ngraph includes
* Remove ngraph rt_info
* Remove unused ngraph includes
* Replace ngraph:: scope with ov:: here and there
* Remove serialize proxy header
* Remove proxy include
* Bring back a file for vpu-plugin
* Fix after upstream merge
* Remove nested namespace conflict
Co-authored-by: Andrei Kochin <andrei.kochin@intel.com>
Co-authored-by: Ivan Tikhonov <ivan.tikhonov@intel.com>
Co-authored-by: Mateusz Tabaka <mateusz.tabaka@intel.com>
* [FE] Fix freezing placeholder via freeze_placeholder_with_value option
Currently, freezing placeholder via freeze_placeholder_with_value option does not work for any frontends. It happens due to absence of a node in _input_shapes dictionary.
* Add get_element_type method for InputModel
* Revert not needed changes
* Revert not needed changes
* Update freeze_placeholder_test
* Add tests for TF FE with freezing values of different types
* Fix Python API return value
* Correct returned type for get_numpy_ctype
* Apply code-review feedback
* Update src/frontends/tensorflow/src/input_model.hpp
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* Apply code-review feedback: no tf legacy specific routine and parameter names
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* Replace primitive's input with input_new and corresponding updates
Signed-off-by: Andrew Park <andrew.park@intel.com>
* unittests updates
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts and remaining unittests update
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Replace node and primitive's deps with deps_new and corresponding updates
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Replace output_padding and output_data_type with output_paddings and output_data_types
- Fix corresponding updates (primitive creation API, calc_output_layout|s, opt passes)
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts and updates
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix failing CI issues
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts and corresponding updates
Signed-off-by: Andrew Park <andrew.park@intel.com>
Signed-off-by: Andrew Park <andrew.park@intel.com>
* [Common] Implement SelectWithOneValueCondition common transformation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add transformation tests for select_with_one_value_condition
* Move call of SelectWithOneValueCondition transformation to MOC
* Apply code-review feedback
* Remove corner case with dynamic output shape for Select
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* update code deps for rhel
* new neo driver version
* new neo driver version fix
* fix
* fix
* drop not needed gpu packages
* removed not necessry NEO packages, updated ov deps
* add gflag for rhel8 os
* update python on ubuntu18 and add core rhel8 rpm URLs
* add libpython3.8
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
* No-break space->space, asssuming->assuming
* benchmark_app: don't print warning each iteration
* Russian с->English c
* Warn no batch only once
* Allow anonymus namespaces
* Keep only one report of model bach size, warn_if_no_batch for static model as well
* [Python API][Tools] Top up upper bound version for dependencies: NumPy, TensorFlow, NetworkX
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Remove upper-bound for TensorFlow and NumPy
* Revert "Remove upper-bound for TensorFlow and NumPy"
This reverts commit 662085df2e.
* Remove upper-bound for NumPy for default installation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Update docstrings after prohibition of not absolute
paths for plugins
* Use only absolute internal paths to load plugins
After introduction of plugins loading via absolute path,
libraries are still tried to load by names if wasn't found.
So safe last place of search (most common) in order to
get useful error message
Co-authored-by: Ilya Churaev <ilya.churaev@intel.com>
* [LPT] FQ Decomposition modified to create FQ without constants with several consumers
* [LPT] Added subgraph test
* CPUTestsBase: Added the ability to check only fusing results
* [CPU] Added subgraph test
* LPT review comments applied
* CPUTestsUtils: added special string 'anytype' for selectedType check ignore
* update code deps for rhel
* new neo driver version
* new neo driver version fix
* fix
* fix
* drop not needed gpu packages
* removed not necessry NEO packages, updated ov deps
* add gflag for rhel8 os
* update python on ubuntu18 and add core rhel8 rpm URLs
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
* serialize gpu memory reuse flag
* added c-tor of reduce_impl
* renamed memory_reuse_by_user as can_reuse_memory, and modified memory allocation logic to be simpler
* [Common] Implement SelectWithOneValueCondition common transformation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add transformation tests for select_with_one_value_condition
* Move call of SelectWithOneValueCondition transformation to MOC
* Apply code-review feedback
* Remove corner case with dynamic output shape for Select
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* sync benchmark Python version with benchmark C++ version.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
* Update tools/benchmark_tool/openvino/tools/benchmark/main.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/utils/utils.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* [AUTO] update property setting logic.
* Update core::set_property() logic to only support primary property for AUTO and MULTI.
* Separate AUTO and MULTI supported properties.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add test case.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. revert the changes for ie core.
2. Enable AUTO/MULTI only accepting its own properties.
3. Enable AUTO/MULTI accepting device properties passed from loadnetwork().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Revert "[AUTO]Update the incorrect config test for Myriad (#13271)"
This reverts commit 0552d98802.
* MULTI only accepts its own properties that is same as AUTO currently.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add test cases for AUTO/MULTI property test.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update to enable MULTI supporting hw primary property setting throw the compile_model().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Remove the added test case for setting secondary property through set_property().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. For AUTO/MULTI, property setting will be passed via core::compile_model() instead of core::set_property().
2. update the logic to infer precision setting that will transform into secondary property setting to each hw device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* set default value for nstreams when -d AUTO/MULTI and no nstreams setting from command line.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update code format.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Enable BA to support if -d AUTO:MULTI,xxx/MULTI:AUTO,xxx. while AUTO Plugin need to update the logic of generating supported config list to virtual device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* move device configuration handling outside form function main.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Revert "move device configuration handling outside form function main."
This reverts commit ef77bfc602.
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Setting property performance hint to 'THROUGHPUT' and passing to executable network if no setting for of AUTO/MULTI.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
* 1. Fix the issue that the dumped configuration file retrieves empty config list when using OpenCV to dump configuration.
2. Enable load_config() to load secondary property.
3. ov:device:properties from command line will replace the ov::device::properties loaded from configuration file.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update the performance statistics showing format.
* Update format.
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update samples/cpp/benchmark_app/main.cpp
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update samples/cpp/benchmark_app/main.cpp
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update README to add some simple configuration files loaded into benchmark app.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. Enable dump/load function for AUTO to support hw device property setting thorugh configuration JSON.
2. Update README.md to example a simple JSON file containing the hw device properties setting for AUTO/MULTI.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
* Update.
* Update.
* Drop the changes within dump/load of OpenCV config parsing.
* Update.
* Correct property name. Update ov::num_stream to ov::inference_precision.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Add attribute for last output element type
* Add convert precision transformation and tests
* Update Unique python API with new attribute
* Update Unique-10 op specification
* Update docstrings
* Update visitor tests
* Add type prop tests for the new attribute
* Check axis constant before shape, compare with partial shape
* Added Debian, Raspbian, Ubuntu 22.10 support
* Updated dependencies.cmake
* Fixes in script
* Added RHEL 9.1 support
* Added proper comments
* Compilation of rpm packages for aarch64 is fixed
* Proper generation of requirements.txt files when ENABLE_WHEEL=OFF
* samples build improvements
* Proper name for json library
* Proper condtion in samples
* [TF FE] Support unknown and undefined Placeholder shape
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Fix mistake in the test
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* added gpu driver version and gpu architecture properties
* deleted unnecessary header inclusion
* reverted to use gfx_ver for architecture string
* device_id is added to caching_properties
* Added BenchmarkLayerTest class for simplifing of benchmarking
* Added example of using BenchmarkLayerTest for GenerateProposals operator
* Added report to file and also updated validate methods
* Switch from using json to xml for BenchmarkLayerTestReporter
* Remove smoke prefix from BenchmarkLayerTest usage example
Co-authored-by: Oleh Kurachenko <oleh.kurachenko@capgemini.com>
Co-authored-by: Irina Efode <irina.efode@intel.com>
* Tests with boolean type for ConvertBroadcast3 transformation
* Extend support of bool input in ConvertBroadcast3 transformation
* New tests redundant ngraph namespace cleanup
* Added checking if imported model version is <= 2.7. Only in such case
custom scale factor will be applied, otherwise there will be only logged
warning message
* Added unit tests for validation of scale factor
* Add more tests for Unique op reference impl
* Use stable sort to preserve elements order
* Update tensor comparator and unify approach for sorted
* Slices comparator revert and test
* Fix slices_ascending_order comparator
* [TF FE] Optimize FusedBatchNorm translator - avoid Transposes
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Fix FusedBatchNorm: correct compute for batch_mean and batch_variance
* Add layer tests and fix based on them
* Mark failing test cases and bind to issue tickets
* Work around tests for FusedBatchNorm in inference mode
* Use separate fictitious constants for reserved outputs
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* [GNA] added dims size validation to avoid segfault in orientation helper
* [GNA] fixed matmul transformation test to not access data from empty container
* [OV] fixed padd Tile to TileIE and added assertion to ngraph to CNN conversion.
* [CONFROMANCE] Fix OV Path in new conformance runner
* filelist
* dd
* try
* one
* Update run_conformance.py
* sss
* revert extra
* f
* Update run_conformance.py
ConvolutionBackpropData weights have following shape {input channels, output_channels, ...},
so unlike forward convolution, the output channels are on the second dimension.
* kernel_impl_params should be serialized before primitive_impl
* reduced the range of min max values for this test case
* updated to use default get_mem when min max are not specified
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. Remove part of AUTO/MULTI test case since they has been coverred in another PR[13738].
2. Move the rest part of AUTO/MULTI test cases into CPU function test suit.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* moved serialization include path
* quiet onednn-gpu patching
* save and load kernels in _impls
* changed to use OPENVINO_ASSERT
* fix errata
* updated to follow OpenVINO naming convention
* updated error messages
* binary buffer by vector<uint8_t>
* partial_shape serialization
* removed object_type
* added a new storage class for primitive_type_string and id
* updated to throw an exception when _node is null in build_deps().
* removed redundant memory_pool clearing
* added a new net_id creator
* newline at eof
* updated CLDNN with GPU
* added cache blob descriptions
* updated output allocation logic in serialization
* added ov::device::architecture in supported properties
* overrided save and load in data_inst and mutable_data_inst
* removed save and load functions in mutable_data
* baseline for serialization unit tests
* added serialization unit tests
* added serialization unit tests
* updated not to execute build_deps when deserialized
* make_data without namespace
* updated to use default layout c-tor
* updated get_unique_net_id()
* updated get_type_id() to a pure virtual method
* updated ov::caching_properties
* added [GPU] tags
* updated network c-tor
* updated unit tests
* Add developer guide for conditional compilation
Change-Id: If769ce1afc045e678b5816ed6ae903ba48a5a320
* Use reference to real code as example
Change-Id: I21f6aa47dfdf4ab7e92424a3ccc3d7524cc4b762
* Add table for macros
Change-Id: I9503d801ebe67ea5c6e525e02a21a7295ba56bc6
* Update for more clear description
Change-Id: I83b39f4a2462fb47ad1537e9747a2bb4a031dab9
* Use refer link replace code snippet
Change-Id: I5e63302718ab02eb4bd7a28ae994c8e5e26e0e02
* Add README.md for CC
Change-Id: I90abe73297deb5f55f219c15b093a02a30f9e8b5
* Update README.md
Change-Id: I8199252cb24fdacba16ea8b9fac63f6f47e4f1e3
* Add link list to doc
Change-Id: Id9a512cf8998cf18795cc0ac096fbbf89b28941c
* Move transformation and lp transformations tests from IeFuncTests and IeUnitTests
* Enable tests in Azure CI
Co-authored-by: Ilya Churaev <ilya.churaev@intel.com>
+ Updated oneDNN to use cache_blob
+ Updated oneDNN to fix group conv failure
+ Add g_os_iyx_osv8 format and relevant reorder to support oneDNN update
+ Used cached_blob to make engine if cache_dir config is used
Signed-off-by: Min, Byungil <byungil.min@intel.com>
* convert() method added.
* Moved conversion to convert() method.
* Fixed commits.
* Output dir fix.
* Added objects support for extesions param.
* Added support for transformations_config extension objects.
* Input to str unit tests.
* Added tests, added comments.
* Updated BOM.
* Removed commented code.
* Fixed extension passing.
* Small corrections.
* Fixed for python 3.6.
* Small fix.
* Moved dir creating to ov.serialize(), removed mo.serialize(), small fixes.
* Small fix.
* Small correction.
* Removed coping of params, moved convert implemetation to separate module.
* Import fixes.
* Moved hiding of exceptions to main().
* Updated comment.
* Fixed unit tests.
* Comment changed.
* Fixed dir creating.
* Tests fixed.
* Small fixes.
* Test fix.
* Added meta data generation, removed printing of execution time for silent mode.
* Import fix.
* Conflict fix.
* Fixed error.
* Fix for custom config.
* Added version, data_type params to help.
* Added mo.convert() full-functional tests.
* Small corrections.
* Comment correction.
* Moved convert to openvino package, moved LayotMap and InputCutInfo to openvino.convert.
* Added help param.
* Wrong change removed.
* Small fix.
* Removed unnecessary comments.
* Removed .xml extension check from append_ir_info.
* Added missed file.
* Fixed error.
* Fix for bool value in InputCutInfo.
* Moved InputCutInfo, LayoutMap to openvino.tools.mo.
* Moved InputCutInfo, LayoutMap to openvino.tools.mo.
* Moved check and read_model to emit_ir.
* Small correction.
* Added comment.
* Added unit_tests with convert().
* Small corrections.
* Removed convert alias from openvino.
* Fixed conflicting unit tests.
* Removed unnecessary warnings.
* Params check fix.
* Small correction.
* Added paths checks.
* Added negative tests for to_str methods, fixed errors.
* Added tuples support in input parameter.
* Added direct support of BytesIO and TF1 graph def.
* Moved reminders to update OV and use API 2.0 to main().
* Fixed keras loading.
* Returned .mapping file generating.
* Added positional input_model param.
* Added test for unnamed input_model.
* Optimize imports.
* Added more informative error for brackets syntax in --input.
* Keras direct model support.
* Keras direct model support.
* Small fix.
* Conflict fix.
* Conflict fix.
* Added direct support of BytesIO and TF1 graph def.
* Fixed keras loading.
* Keras direct model support.
* Keras direct model support.
* Small fix.
* Tests for importing from memory.
* Support of other tf/pytorch formats.
* Updated mo_convert extensions tests to use ov models.
* Removed debug output.
* Implemented PyTorch converting logic.
* Small corrections.
* Small corrections.
* Added comments.
* Fixed for single input case.
* Added switching between save to file and BytesIO.
* Small fixes.
* Rename convert() to convert_model().
* Added env variable to disable converting to onnx.
* Tests refactoring.
* Add MO Python API tests to precommit.
* Add MO Python API tests to precommit.
* Added PyTorch to layer tests requirements.
* Added supported formats description.
* Fixed errors, added tests.
* Fixed bugs, added support of numpy and ov.Tensor sample_input.
* Added more torch.Size tests.
* Small correction.
* Renamed sample_input->example_inputs.
* Tests refactoring.
* Code style.
* Added support of dict in example_inputs.
* Small correction.
* Added removing of tmp onnx model in case of conversion error.
* Fix for lists of tensors in example_inputs.
* Support of dynamic axes for Keras layer, Keras module.
* Removed disabling of eager execution.
* Tests fixed.
* Added pytest requirement for layer tests.
* Added convert_model to openvino.runtime, fixed test runs.
* Small fix.
* Tests fix.
* Better help message.
* Error fixed.
* Renamed example_inputs->example_input.
* Small fix.
* Added support of list of layouts.
* Removed wrong change.
* Added tuple support for --layout.
* Removed convert_model from openvino.runtime.
* Use of default onnx opset version.
* Added support of dynamic input shapes without example_input.
* Made better error message.
* Removed stack trace from exceptions.
* Fixed tests.
* Small fix.
* Removed wrong change.
* Added import model from memory tests in unit tests.
* Replaced compare_functions() with MO IR reader compare.
* Removed test.
* Removed not needed change.
* Fixed conflicts.
* change gpunum to 3
* hold threads for GPU for MULTI:GPU,CPU
* need to first check if there is a CPU in the device list
* use getNumberOfCPUCores to get CPU cores
* load GPU first
* assign the correct value to multiSContext->_devicePriorities
* load GPU first and load CPU last and set numa for CPU
* MULTI set CPU affinity to “NUMA” during load network
* Load the CPU last while maintaining the original device priority
* not using vector for CPU
* There is no user setting affinity in MULTI, and NUMA is set for the CPU
* pass key ENABLE_HYPER_THREAD to CPU plugin and merge xiaoxia PR
* set ENABLE_HYPER_THREAD to NO
* modify log
* Modify the code according to xiaoxia and wanglei comments
* Modify the code according to bell comments
* [TF FE] Break the cycle for NextIteration
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Fix code-style
* Better to cut by outputs of operation creating the cycle
* Remove extra thing
* Add test model
* Add test for conversion of TF1 While
* Apply code-review feedback: use pointers and correct error message
* Remove extra check
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* fix custom setting nstreams doesn't work due to the logic changed in ApplyPerformanceHints
* add same limitation for latency mode
* change bind core method to binding full physical core first, then logical core
* update custom setting streams and threads
* fix corner case when threads_per_stream is 1
* fix with comments
* add condition of big core stream is zero when binding core
Co-authored-by: Shen, Wanglei <wanglei.shen@intel.com>
Co-authored-by: Chen Peter <peter.chen@intel.com>
It might not be available in some cross-toolchains.
Extract the code for switching to gold linker into separate
CMake utility function to avoid duplication between multiple CMake scripts.
* Moved MO meta data to RT info.
* Undo wrong change.
* Reformat code.
* Code reformat.
* Corrected error message.
* Test correction.
* Test fix.
* Updated MO IR Reader.
* Added testing of meta data for MO IR reader.
* Code corrections.
* Default params removed from meta data.
* Test corrected, fixed paths cropping.
* Small fix.
* Renamed legacy_path to legacy_frontend.
* Fixed error.
* Added PartialShape from str constructor, added to_string() method.
* Small corrections.
* Moved shape parsing to PartialShape and Dimension classes.
* Fixed errors.
* Added tests for shapes with comma separator.
* Fixed tests.
* Test fixed.
* Fixed test.
* Fixed tests.
* Small fix.
* Removed usage of depricated method.
* Fix in MO IR reader.
* Tests fixed.
* Fix in MO IR reader.
* Test fixed.
* Shape class alignment.
* Code fixes.
* Code fixes.
* Used unnamed namespace.
* Undo change of Shape attribute adapter.
* Undo change of Shape attribute adapter.
* Removed unnecessary includes.
* Removed unnecessary code.
* Test fixed.
* Fix in MO IR reader.
* Moved check_all_digits() to ov::util.
* Moved check_all_digits() to src/core/src/.
* Clang format.
* Apply suggestions from code review
Co-authored-by: Ilya Churaev <ilyachur@gmail.com>
* Removed dimension_util file.
* Removed wrong change.
* Removed extra spaces from string representation of Shape.
* Test fixed.
* Added support of comma separator for value. Updated help.
* Returned old format shape serializing.
Co-authored-by: Ilya Churaev <ilyachur@gmail.com>
* Basic Unique impl
* Shape inference tests
* More type prop tests
* Dynamic input tests
* Visitor tests
* Code formatting
* Minor corrections
* Modern c++ trick to make MSVC happy
* Refactor
* Code review
* More tests with dynamic shapes
* Op check test for Unique
* Well, a whitespace
* Different way of out shape calculation
* Spec
* Axis values constraint in the docs
* Zero-dim handling
* opset10_dump test adjustment
* 0D implementation and some test infra adjustments
* More coverage for scalar inputs
* type dependent test case generators
* A working version for 1D with all outputs
* Tensor shape calculation helper
* 1D handling with sorted mode included
* Handling of no-axis cases
* Some tests for no-axis case
* Unique's evaluate method
* A working fallback for the CPU plugin
* evaluate() removal
* Corrected shape inference
* More type prop fixes
* Unified axis extraction helpers
* Cleanup and preparation for slices processing
* Slice descriptor struct
* Static shapes generation with axis
* N-D solution with axis handling
* Slices comparison impl
* Fix Windows compilation errors
* Apply defaulting principles
Co-authored-by: Tomasz Jankowski <tomasz1.jankowski@intel.com>
* [GPU] Move device memory allocation from preprocessing stage to runtime to fix incorrect behaviour in case of tensor swapping
* [GPU] Apply review comments
* Replace source of dynamic layout cheking
* [AUTO] update property setting logic.
* Update core::set_property() logic to only support primary property for AUTO and MULTI.
* Separate AUTO and MULTI supported properties.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add test case.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. revert the changes for ie core.
2. Enable AUTO/MULTI only accepting its own properties.
3. Enable AUTO/MULTI accepting device properties passed from loadnetwork().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Revert "[AUTO]Update the incorrect config test for Myriad (#13271)"
This reverts commit 0552d98802.
* MULTI only accepts its own properties that is same as AUTO currently.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add test cases for AUTO/MULTI property test.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update to enable MULTI supporting hw primary property setting throw the compile_model().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Remove the added test case for setting secondary property through set_property().
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. For AUTO/MULTI, property setting will be passed via core::compile_model() instead of core::set_property().
2. update the logic to infer precision setting that will transform into secondary property setting to each hw device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* set default value for nstreams when -d AUTO/MULTI and no nstreams setting from command line.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update code format.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Enable BA to support if -d AUTO:MULTI,xxx/MULTI:AUTO,xxx. while AUTO Plugin need to update the logic of generating supported config list to virtual device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* move device configuration handling outside form function main.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Revert "move device configuration handling outside form function main."
This reverts commit ef77bfc602.
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Setting property performance hint to 'THROUGHPUT' and passing to executable network if no setting for of AUTO/MULTI.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
DisableConvertConstantFoldingOnConstPath was renamed to MarkDequantizationSubgraph
and modified such that it also marks Subtract and Multiply from dequantization subgraph
with 'DequantizationNode' attribute.
Ticket: 65489
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
* [GPU] Add weight format to enable models
Add g_os_zy_is_x_osv8_isv2, g_os_zyx_is_osv8_isv2, g_os_is_zyx_isa8_osv8_isv2
These formats are required from brain-tumor-segmentation-0001/mxnet/fp16/bs1,
resnext-101-3d/onnx/fp16/bs1.
* [GPU] Fix the issue, Can't find reduce output later in OutputsDataMap.
The layer name of Reduce converted from AvgPool is different originals.
It caused an exception, so use original name.
We have AvgPool to Reduce conversion to improve Perf in dGPU.
-target model
action-recognition-0001-encoder, driver-action-recognition-adas-0002-encoder
* [GPU] Remove bfzyx format in oneDNN FC supported format
Current clDNN FC supports 4dims NOT 5dims, and many related logics are implemented from this, likes
calc_output_layout.
For this reason, oneDNN FC 5dims occur the functional issue in resnext-3/onnx/FP16/BS1
* [GPU] Fix to unsupported 6dims block-format in gather cals_output_layout.
It's failed issue in cocosnet/onnx/FP16/BS1
Signed-off-by: hyunback <hyunback.kim@intel.com>
* ov::set_property should not set HW plugin when device is AUTO or MULTI plugin.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* core set_property() didn't support setting ov::device::propreties for HW device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Added doxygen comment for opset10
* Added get_available_opsets method to API
* Fixed build
* Fixed build and typo
* Use constant map
* Removed defines from global opset
* Added const to map
* turn on bom_tests
* fixed root directories to search
* leftovers fix
* removed requirements from package_BOM.txt; corrected error messages to install via pip
* removed extract_release_version.py from package_BOM.txt, added requirements files to skip
* removed redundant commented line
* mentioned CustomLayersMapping.xml.example in docs
Co-authored-by: Andrei Kochin <andrei.kochin@intel.com>
* Intial update for NMS mutiple outputs support
- update get_depedencies_new primitive overriding func for nms and reorder
- primitve API update for nms and reorder
- add non_max_suppression_base test TC for multiple outputs
- update non_max_suppression cpu impl for backward compatibility
- update dependencies_new for remove_redundant_reorders opt pass
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Update program_node and kernel_impl_params for multiple outputs supports
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix runtime error and failed TC for multiple outputs support
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts and newly update
Signed-off-by: Andrew Park <andrew.park@intel.com>
* TopK shape infer
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Implement TopK gpuFunctests for dynamic shape and multiple outputs
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts
Signed-off-by: Andrew Park <andrew.park@intel.com>
* minor API fix
Signed-off-by: Andrew Park <andrew.park@intel.com>
* NMS shape infer
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Implement Nms gpuFunctests for dynamic shape and multiple outputs
Signed-off-by: Andrew Park <andrew.park@intel.com>
* move gpuFunctests and fix conflict
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix issues for Nms gpuFunctests
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts and code cleanup
Signed-off-by: Andrew Park <andrew.park@intel.com>
* apply clang-format
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply change to onednn primitive_impl
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix CI issues
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
* updates for ngraph shape infer
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix conflicts
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Failing case
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
Signed-off-by: Andrew Park <andrew.park@intel.com>
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
* [TF FE] Refactor translator for L2Loss and test in the pre-commit
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Use Power for square operation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* [PYTHON] Pipeline transfer
* [PYTHON] Align python benchmark
* [PYTHON] Align last step
* [PYTHON] Fix innacuracies of the last step - median
* [PYTHON/CPP] Add Core::get_version method to python API, refactor Ben benchmark to print version with this func
* [PYTHON] Remove get_version_string from API
* [PYTHON/CPP] Align output for model input/output info
* [PYTHON/CPP] Step 4,6 alignment of outputs, step 8 dumps all info stored in config parameters
* [CPP] Fix a bug causing nstreams parameter to never be set to AUTO in CPP benchmark_app
* [CPP] Fix clang format errors
* [CPP] Modify print order and data output for 8th step
* [PYTHON] Add verification checks from C++, modify set_thoughtput_streams to match documentation
* [CPP] Revert changes to C++ benchmark_app
* [CPP] Remove additional spacebar
* Update submodules versions on remote
* Update module from master branch
* Redownload submodules from master and override changes from commit
* [PYTHON] Remove unneccesary parse_status from validation function
* [PYTHON] Check for HINT in map, fix circular import
* [PYTHON] Remove artifacts from commit, fix args.perf_hint set to '' instead to 'none'
* [PYTHON] Reverse changes to perf hint, add key in map check, fix validation function throwing error on set hint
* [PYTHON] Fix linter
* [PYTHON] Remove linter spacebar
* [CPP] Fix wait_all exception throw
* [CPP/PYTHON] Clean artifacts and unwanted changes from work process
* [PYTHON] Fix artifacts from merge, clean submodule update
* [C++ CPU] Fix device name string by removing padding NULL characters from the back
* [CPP] Fix ba infer_request_wrap in other throw-catch clauses
* [PYTHON/CPP] Fix missing latencies in final step for shape group, fix minor misaligned messages, add missing report parameter create infer requests time
* [CPP] Clang fix formatting
* [CPP] Reverse clang fix format on plugin.cpp
* [PYTHON/CPP] Fix C++ progressbar printing endl when disabled, fix rounding in python creating infer request message
* [CPP] Fix foramtiing error
* [PYTHON/C++] Refactor network to model based on naming conventions, provide fresh README output example
* [PYTHON/C++] Add example output to C++ README, remove unnecessary device loop
* [BENCHMARK_APP/C++] Fix artifact from refactoring, remove try-catch clause
* Update samples/cpp/benchmark_app/benchmark_app.hpp
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update samples/cpp/benchmark_app/main.cpp
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/main.py
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update samples/cpp/benchmark_app/main.cpp
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* [CPP] Fix clang errors
* [CPP/PLUGIN Reverse modification to extract to separate task
* Update tools/benchmark_tool/openvino/tools/benchmark/main.py
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/parameters.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/utils/utils.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Update tools/benchmark_tool/openvino/tools/benchmark/main.py
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* [PYTHON/C++/BENCHMARK_APP] Fix language inconsistencies, remove unnecessary checks
* Update pyopenvino.cpp
* [CPP/BENCHMARK_APP] Remove unnecessary try-catch, fix linter errors
* [PYTHON/CPP/BENCHMARK_APP] Revert changes to Core, align version prints usin only provided methods
* [DOCS/BENCHMARK_APP] Update README with proper model examples
* Update README.md
Co-authored-by: Nadezhda Ageeva <nkogteva@gmail.com>
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* [GPU] Refactor impls factory method
* [GPU] Return unique_ptr from impls factory
* [GPU] implemented get_kernel_params for most of ocl impls
* [GPU] Common create impl
* [GPU] Fixes after rebase
* [GPU] Codestyle fixes
* [GPU] handle optimized out prims in common create
* [GPU] Fix exception type in region yolo tests
* [GPU] Resolve ambiguity of make_unique on windows
* [GPU] Multiclass NMS
* [GPU] Fixes after rebase
* gpu graph serialization
* fix to rebase
* onednn_gpu.patch for serialization
* git apply --verbode to --quiet
* functional tests
* removed referece of mas_unpooling.hpp
* git apply --verbose
* add no args ctor for proposal_impl
* changed kernel_cache save/load error messages
* gpu model cacning control env. variable
* fixed nonnull warning
* impl_params are added to save and load
* changed a way to use kernel_impl_params in save and load
* get_arguments_by_idx is added
* setenv is disabled in windows
* added missed part for onednn
* code refactoring based on code review
* fixed to use get_node_output_layout()
* OV_GPU_MODEL_CACHING is changed to OV_GPU_CACHE_MODEL
* reference to node and primitive are removed
* fixed hash of enum class
* restored CanShareContextWith
* serialization of intermediate memory
* fix to rebase
* multiclass_nms serialization
* caching_properties added
* add HybridAware stream setting and core binding
* fix clang format issue
* unified code style, add parameter check
* correct input affinity skip NUMA, modify function name
* remove unnecessary floor
* fix ci compile issue on Mac/windows platform
* modify smoke_SetConfigAffinity test
* modify ov_core_set_and_get_property_enum test, affinity HYBRID_AWARE is changed to NUMA
* remove affinity correcting on this pr
* revert ov_core_test.cpp
* merge function by comments
* fix code style issue
* add custom nstreams setting, remove mutable qualifier
* fix code style issue
* fix some issues according to comments
* modify UpdateHybridCustomThreads
* fix code style issue
* modify comments
* use logic core in phase 2
* modify steams threads setting method
* add custom setting nstreams nthreads
* modify comments
* fix code style issue
* fix threads_per_stream_big is wrong in the some ADL machine which has 6 pcore
* add one condition that custom set nthreads but not set nstreams
* remove the limit to nstreams
* remove the feature in branch p_e_core_phase2_logic
* add comments
* rename num_small_cores to num_small_cores_phys
* fix code style issue
* fix thread number is wrong in corner case on 6Pcore and 10Pcore condition
* fix some condition which nstreams<2
* add check for config._threads_per_stream_big
Co-authored-by: Shen, Wanglei <wanglei.shen@intel.com>
* Blocked layouts support for ExperimentalDetectronGenerateProposalsSingleImage
* fix RHEL 8.2 build
* reduce number of tests
* Added bs_fs_yx_bsv16_fsv32 format
* [PYTHON] CV2 module required only when loading images
* [PYTHON] Update tools/benchmark_tool/openvino/tools/benchmark/utils/inputs_filling.py
Co-authored-by: Helena Kloosterman <helena.kloosterman@intel.com>
* [PYTHON] Make CV import not global
* [PYTHON] Update format of inputs_filling.py
* [PYTHON] Fix missing newline
* Update tools/benchmark_tool/openvino/tools/benchmark/utils/inputs_filling.py
Co-authored-by: Helena Kloosterman <helena.kloosterman@intel.com>
* [BENCHMARK_APP/PYTHON] Introduce exception when module is installed but fails to import
* [SAMPLES/PYTHON] Sporadic access violation fix through checking whether module was cached
Co-authored-by: Helena Kloosterman <helena.kloosterman@intel.com>
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
* Move GeluFusion into ov namespace
It required to move utils as well.
* Move AddFakeQuantizeFusion into ov namespace
* Move AlignEltwiseInputRanks into ov namespace
* Move BatchToSpaceFusion into ov namespace
* Move BinarizeWeights into ov namespace
* Move BroadcastElementwiseFusion into ov namespace
* Move ClampFusion into ov namespace
* Move Concats and Eliminations into ov namespace
* Move many passes and such into ov namespace
* Move many passes and such into ov namespace
ngraph_ops
op_conversions
* Move control_flow into ov namespace
* Apply style
* Move smart_reshape into ov namespace
* Move opset_conversions into ov namespace
* Move low_precision into ov namespace
* Apply style
* Move fused_names into ov namespace
* Add missing aliases
* Move few passes into ov namespace
ConvertPrecision
FixRtInfo
InitNodeInfo
* Move CommonOptimizations into ov namespace
* Fix convert_precision conflicting namespaces
* Apply code style
* ov-nize Type Relaxed definitions
* Remove useless aliases
bulkly added by script
* Revert opsets
* Address review comments
* Remove NGRAPH_RTTI
* Clean up remains of yet another implicit inclusion
Co-authored-by: Andrei Kochin <andrei.kochin@intel.com>
Co-authored-by: Ivan Tikhonov <ivan.tikhonov@intel.com>
* Single-layer test:
- support both v8 and v9 in shared test;
- add test for GPU.
* Transformation:
- convert both v8 and v9 to internal operation;
- add transformation to GPU pipeline.
* GPU primitive and CreateOp() for MulticlassNMS operation
* kernel stuff
* add unittests
* Added bs_fs_yx_bsv16_fsv32 format
* add int64 support
Co-authored-by: Mykhailo Hnap <mhnap@lohika.com>
* Set db_api_handler for pushing to new database in timetests
* Change function of pushing to db.
Co-authored-by: Daria Ilina <daria.krupnova@intel.com>
* Handle DT_UNKNOWN as DT_REG during file list-up
Some filesystem does not return dt_type properly. (such as sshfs)
Without this fix, openvino fails to search frontend when mounted on
sshfs.
* style fix
* style fix
* [GPU] Added unit tests for roi polling operations.
* [GPU] Added blocked layouts support for roi_pooling operations
* [GPU] Added single layer tests for roi polling operations.
* [GPU] Fix attach roi pooling impl, remov const from formats and types
* [GPU] Fix roi polling kernel, bilinear pooling operation
* Fix, whenever output_width and output_hight are equal to 1, which cause a dividing to 0.
* Fix, for proper calculation y and x.
Co-authored-by: Mykhailo Hnap <mhnap@lohika.com>
* Add reference implementation for Is ops
* FIx test
* Fix typo
* Remove deprecated
* Fix issues with API switch
* Force cast to float for float16/bfloat16 in ref
* Fix style
* Remove BWDCMP_RTTI
* Remove TensorVector helper check
* Remove evaluate functions
* Disable faling tests due to missing eval
Removing the 'global' tabs and preparing a language-agnostic version of the article. Replacing png image with a scalable svg file. Proofreading the article.
* [Test] Build doc for C API reperence
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [Test] Build doc for C API reperence add C path
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [Test] Build doc for C API reperence add C path
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API][DOC][DOXYGEN] refact the components of C API
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [DOC][API reference] remove useless code
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [C API][DOC] Modify format to generate api reference
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [DOC][API reference] fix doc build error
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* Enable linker warnings as error for Windows
* Fisable some linker warnings
* Suppress warnings
* Fixed build
* Change if opset
* Try to fix some warnings
* Revert some changes
* Try to fix warning
* Define OpenVINO static library for static libraries
- Separate kernels_cache::add_kernel from factory (choose_impl)
- Reset kenrels_cache._kernels after kernels_cache.compile
- Create cldnn unit test case to check multi-stream processing
* [TF FE] Refactor SpaceToBatch and BatchToSpace operations
Make them to support inputs of dynamic rank and shapes
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add layer tests in the pre-commit
* Fix for normalizing padding and crops
* Correct the type of paddings and crops
* Fix incorrect changes for paddings and crops
* Update src/frontends/tensorflow/src/op/space_to_batch_nd.cpp
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* Update src/frontends/tensorflow/src/op/space_to_batch_nd.cpp
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
* Add ctest for some test targets
* Fixed test coverage
* Replaced unit tests cmake script
* Fixed code style
* Use ov_add_test_target for FEs
* Fixed code style
* Enabled Conv dynamic shape & Added conv dynamic functional test
* Fix layout optimizor
* Revert layout optimizer change
* Set expected layout of convolution for dynamic shape to be fsv16
* Fixed groupConv for dynamic shape
Added 1D dw/group convolution tests for both static & dynamic
Currently, 1D group convolution is only works for dw..
* Minor fix
* Updated for 3D conv dynamic case + corresponding test
* Fix lint error
* Fix INSTANTIATE_TEST_CASE to INSTANTIATE_TEST_SUITE
* Fix clang format & skip 1D group conv test
+ Added logic to select planar format for b-f axes reduction
+ Bugfix of reorder_unreduced_axis_no_fusion
+ Bugfix to resolve an issue of cgan_mnist
Signed-off-by: Min, Byungil <byungil.min@intel.com>
* Update the logic to handle the ov:device::properties in the ie core.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update test cases.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* 1. Remove the config filter within AUTO plugin when passing the configs to target hardwre device through AUTO plugin.
2. add test cases to check if the secondary properties have been passed into the target device through virtaul device.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add mock tests to check if device properties works throgh the AUTO or MULTI plugin.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Add mock head file for loadnetwork with device properties testing.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* Update gtest commit to fix the warning issue appeared in the macro MATCHER_P.
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
Signed-off-by: Wang, Yang <yang4.wang@intel.com>
* [TF FE][Common] Move TF FE to opset Unique and add Unique decomposing to common
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add Unique decomposition transformation into the common part
* Add transformation tests for UniqueDecomposition
* Fix unique_decomposition_test
* Simplify condition of transformation applicability
* Clarify the comment when the transformation is applicable
* Fix transformation test
* Fix transformation test
* Add support of opset10 in IR Reader
* Mark the single element case as xfail
* Fix tensor names in the transformation
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Remove py3.6 constraints
* Cleanup out of date python versions
* Bump pybind11 to v2.10.1 release tag
* [PyOV] Fix wheel requirements for new package versions
Co-authored-by: jiwaszki <jan.iwaszkiewicz@intel.com>
* Add shape_infer function for GRUCell op
* Add shape_infer function for AUGRUCell
* Consts refactor
* Add batch_size check
* Enable GRUCell shape_infer for CPU
* Style apply
* Use OV_EXPECT_THROW in tests
* Use helper for input shapes
* Use .back() instead of index
* Change rnn_seq namespace to rnn
* Fix win warnings
* Enable AUGRUCell/Sequence shape_infer on CPU
* Fix warn
* Fix warn
* Review squeeze operator for
- label preserve and propagation
- partial values preserve and propagation
- interval shapes propagation
* Review static shape inference tests
* Add template shape inference for squeeze
* Update include for OV_EXPECT_THROW
* Correct has_static_axes initialization
* remove test_squeeze from xfail 44968
* Fix inference parameter op with no data
- add additional test to check if axes are parameter without data
- revert xfail marks for squeeze tests
* Fix inference parameter op with no data
- add additional test to check if axes are parameter without data
- revert xfail marks for squeeze tests
* Refactor axes acquire logic
* Enable TLV
* [GNA] Add smoke test for Conv with padding on MTL
* [GNA] Allow skipping tests based on gna lib version
Co-authored-by: Krzysztof Bruniecki <krzysztof.bruniecki@intel.com> 🥇
After fixes for TopK and arithmetic operations for integer tensors, we are ready to switch
these tests in the pre-commit
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Added GNA logger
* Added GNA logger class
* Replaced old looger by the new one
* Added support of the all log levels in the tests
* Renaming
* fixed cpplint issue
* Reduced error level
* fixed cpplint issues
* Fixed linking error
* Fixed linux build
* Rebase fix
* Removed unused code
* Fixed compile issue after rebase
* Aligned code usage and style
* Fixed default constructor
* Fixed test
* One more rebase
* Added exveption test
* Fixed the old logging tests
* restored static constants
* Removed dublicated log messages
* Removed incorrect log::level comparation
* Fixed generic api test
* Rebased
Current xbyak has "unresolved external symbols" issue
when built on Windows in 'Debug' (-O0) build type
The issues is fixed in the recent xbyak versions
# FIXME: there are compiler failures with LTO and Cross-Compile toolchains. Disabling for now, but
# this must be addressed in a proper way
ie_dependent_option(ENABLE_LTO"Enable Link Time Optimization"OFF"LINUX OR (APPLE AND AARCH64);NOT CMAKE_CROSSCOMPILING;CMAKE_CXX_COMPILER_VERSION VERSION_GREATER 4.9"OFF)
ie_dependent_option(ENABLE_LTO"Enable Link Time Optimization"OFF
"LINUX OR (APPLE AND AARCH64);NOT CMAKE_CROSSCOMPILING;CMAKE_CXX_COMPILER_VERSION VERSION_GREATER 4.9"OFF)
ie_option(OS_FOLDER"create OS dedicated folder in output"OFF)
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
