* Use Serialization as a default engine in MO
* Added cmd option to use old serialization
* Added mapping file generation
* Test mapping file generation
* Fix setBatchsize parameters order; fix mapping file generation
* Added FrameworkNode; added method to read models with custom ops but without extensions
* Added python API for read_network_without_extensions function; updated mo not to use IECore
* Added read_model_without_extensions to IReader and IParser
* Fix V7 IR reader
* Fix pword value
* Fix dllexport macro usage
* Add metainfo to IR
* Fix nGraph code style
* Fix license header
* Restore prepare_emit_ir behaviour
* Fix compare_function to resolve situation when Result input port has multiple names
* Update Compare Functions
* Fix FrameworkNode validation
* Self-review
* CodeStyle check
* --use_fallback -> --use_legacy_ir_generation
* Sort imports in main.py
* --path_to_model -> --input_model
* Use logging instead of print
* Code simplifucation&cleanup
* Fix offline_Transformations key
* Fix GeneraeMappingFile comments
* Use Extension approach to work with custom ops
* Fix versions check
* Code clean-up
* Moved FrameworkNode to inference_engine_transformations library
* Fix FrameworkNode includes
* Code clean-up
* Extend blobs dumping with filtering by environment variables
The idea is to dump blogs without rebuilding
We cannot just use environment variables without compile-time flag
because of security flaws.
Instead, it is expected that developers just always set additional
macro (BLOB_DUMP_PATH) which is not set for the production builds:
export CXXFLAGS="-DBLOB_DUMP_PATH=\\\"mkldnn_dump\\\""
This macro activates blob dump filtering using environment variables.
To prevent unnecessary dumping, blobs are not dumped by default even
if macro is defined.
* Extend nGraph Python API and test IE IR reader for Einsum
Signed-off-by: Roman Kazantsev <roman.kazantsev@intel.com>
* Format description for test auxiliary function
Signed-off-by: Roman Kazantsev <roman.kazantsev@intel.com>
* Remove print from the python test
Signed-off-by: Roman Kazantsev <roman.kazantsev@intel.com>
* [cldnn] Add initial fused conv eltw POC
- Add cldnn unit test
- Add fused dependency list to the fused_primitive_desc
- fuse_nodes update for saving fusing history and depenecies
- Modify Jitter to create jit constants using fused dependencies
- Add cldnn unit-test cases for multiple serial and parallel eltwise fuse pattern
- Modify Jitter and add default values in sum input
Signed-off-by: Ahn, Paul Y <paul.y.ahn@intel.com>
Co-authored-by: Andrew Kwangwoong Park <andrew.kwangwoong.park@intel.com>
* [cldnn] Update fused_conv_eltwise cldnn unit test
- Add execute and compare function
- Add cldnn unit-test case for multiple parallel eltwise and additional eltwise
- Add cldnn unit-test case for combination of multiple parallel eltw
- Add cldnn unit-test cases for serial and diverged quantize and eltwise
Signed-off-by: Andrew Kwangwoong Park <andrew.kwangwoong.park@intel.com>
* [cldnn] Modify checking fusibility of eltwise fusing
- Add new checking fusibility rule in prepare_primitive_fusing
- Move cldnn eltwise fusing test to fusing_gpu_test.cpp
- Modify method to get input var name in jitter
Signed-off-by: Ahn, Paul Y <paul.y.ahn@intel.com>
* [cldnn] Fix fusing item type and activation fusibility checking condition
- Extract input_data_supports_fusings from fuse_activaion_f
- Fix checking supported mode bug
Co-authored-by: Andrew Kwangwoong Park <andrew.kwangwoong.park@intel.com>
* Fixed preprocessing for template plugin
* Added more tests instances
* Split common transformation to smaller ones which can be used by plugins
* Moved preprocessing transformation to Plugin API
* Added PreprocessConversionTest tests
* Disabled tests on GPU: CVS-51764
* Disabled some tests on VPU and TEMPLATE
* Support for input layout conversions in TEMPLATE plugin
* Improvements in Template Plugin
* Fixed compilation
* Fixes
* Disables some tests
* Fixed compilation on Windows
* Fixed docs
* Small refactoring in TEMPLATE plugin
* Fixed compilation on Windows
* Fixed code style
* Hide CALL_STATUS_FNC helpers to private API
* Moved some base classes to private place from plugin_api
* Updates for VariableState creation
* Take Jane's changes for Demension names
* Revert "Take Jane's changes for Demension names"
This reverts commit 9f6c8fa5a6.
* Removed ICNNNetwork include
* removed more icnnnetwork includes
* Added missed include with ie_input_info.hpp
* Fixed GNA plugin to provide names w/o \0
In linux, without malloc_trim, some freed memories are not being returned to system.
Current hypothesis is that a large allocation for compilation is not completely freeed, thought mostly freed.
This does not happendin Windows.
So, added malloc_trim for linux build until we figure out a better solution.
* initial changes (IR not generated)
* extractor fix
* convert tdnnf (with correct infer)
* refactoring + comments in code
* added unit tests + couple fixes based on tests
* change order for old convolutions
* fix pylint
* small refactoring
* added if to remove changes in old irs
* doc updated
* fix layout and kernel shapes for old convolutions
* fixed test
* moved test
* fix import in test
* fixed backward compatibility
* review fixes
* [PP] FLuid level tests for mean value preprocessing
* PP] Fluid operations for mean value preprocessing
* * Relaxed tolerance and fix for issue.
* * Fix for issue.
* * Applied comments.
Co-authored-by: Anton Potapov <anton.potapov@intel.com>
* Compile time enabling or disabling of first inference time counters
* First inference time counters
* Counters for validate_nodes_and_infer_types and check_all_parameters_registered removed from first inference time counters scope
* Code style fix
* Missing macro for CC and invalid domain names
* Code style fix
* Unused function warnings fixed
* Make find_package(TBBBIND_2_4) silent + Remove inheritance from tbb::task_arena for custom::task_arena
* Make my_binding_observer unique_ptr
Co-authored-by: Kochin, Ivan <ivan.kochin@intel.com>
* Update attributes info
* Update math form to render sqrt symbol
* Update input description
* Update output description
* Add new example
* Update description and move math formula
* Style refactor
* Add info about integer rounding and negative values
* Update examples with integer cases
* Enhance dynamic shape inference and node checks of validate and infer types method
* Remove type prop unit tests from convolution file
* Add type prop unit tests covering dynamic shape scenario
* Allow integer types in validate and infer types method
* Add output shape inference tests with output_padding attribute
* Fix comparison of integer expressions of different signedness compilation error
* Fix code style
* change the deprecated method to the recent
* first ver of the hybrid cores aware CPU streams (+debug info)
* more debug and fixed sum threads
* disabled NUMA pinning to experiment with affinity via OS
* further brushing of stream to core type logic
* hybrid CPU-aware getNumberOfCPUCores
* adding check on the efficiency
* experimental TBB package (that cmake should pull from the internal server)
* iterating over core types in the reversed order (so the big cores are populated first in case user specified less than all #threads)
* adding back the NUMA affinity code-path for the full validation (incl 2 sockets Windows Server)
* cpplint fix and tabbing the #if clauses for the readbility
* pre-production TBB from internal server
* wrapping over #cores/types
* wrapping over #cores/types, ver 2
* wrapping over #streams instead
* disabling warnings as errors for a while (to unlock testing)
* accomodating new TBB layout for dependencies.bat
* next tbb ver (with debug binaries that probably can unlock the commodity builds, without playing product_configs)
* minor brushing for experiments (so that pinning can be disabled)
* minor brushing from code review
* Updating the SHA hash which appeared when rebasing to the master
* WIP refactoring
* Completed refactoring of the "config" phase of the cpu stream executor and on-the-fly streams to core types mapping
* making the benchmark_app aware about new pinning mode
* Brushing a bit (in preparation for the "soft" affinity)
* map to vector to simplify the things
* updated executors comparison
* more fine-grained pinning scheme for the HYBRID (required to allow all cores on 2+8 1+4, and other LITTLE-skewed scenarios)
TODO: seprate little to big ratio for the fp322 and int8 (and pass the fp32Only flag to the MakeDefaultMultiTHreaded)
* separating fp32 and int8 intensive cases for hybrid execution, also leveraging the HT if the #big_cores is small, refactored. also switched to the 2021.2 oneTBB RC package
* code style
* stripped tbb archives from unused folders and files, also has to rename the LICENSE.txt to the LICENSE to match existing OV packaging tools
* assigning nodeId regradless of pinning mode
* tests OpenCV builds with same 2021.2 oneTBB, ubuntu 18/20
* cmake install paths for oneTBB, alos a ie_parallel.cmake warning on older ver of TBB
* Updated latency case desc to cover multi-socket machines
* adding centos8 OCV with oneTBB build
updating TBB drops with hwloc shared libs added.
* enabled internal OCV from THIRD_PARTY_SERVER to test thru CI..
Added Centos7 notbb OCV build (until g-api get ready for onetbb) to unlock the Centos7 CI build
* separate rpath log to respect one-tbb specific paths
* fixed SEQ code-path
* fixed doc misprint
* allowing all cores in 2+8 for int8 as well
* cleaned from debug printfs
* HYBRID_AWARE pinning option for the Python benchmark_app
* OpenVINO Hybrid CPUs support
* Remove custom::task_arena abstraction layout
* Get back to the custom::task_arena interface
* Add windows.h inclusion
* Fix typo in macro name
* Separate TBB and TBBbind packages
* Fix compile-time conditions
* Fix preprocessors conditions
* Fix typo
* Fix linking
* make linking private
* Fix typo
* Fix target_compile_definitions syntax
* Implement CMake install logic, update sha hash for the tbbbind_2_4 package
* Add tbbbind_2_4 required paths to setup_vars
* Update CI paths
* Include ie_parallel.hpp to ie_system_conf.cpp
* Try to update dependencies scripts
* Try to fix dependencies.bat
* Modify dependencies script
* Use static tbbbind_2_4 library
* Remove redundant paths from CI
* Revert "cleaned from debug printfs"
This reverts commit 82c9bd90c5.
# Conflicts:
# inference-engine/src/inference_engine/os/win/win_system_conf.cpp
# inference-engine/src/inference_engine/threading/ie_cpu_streams_executor.cpp
# inference-engine/src/mkldnn_plugin/config.cpp
* Update tbbbind package version
* fixed compilation
* removing the direct tbb::info calls from CPU plugin, to aggregate everything in the single module (that exposes the higher level APIs)
* Update tbbbind package version
(cherry picked from commit f66b8f6aa6)
* compilation fix
* brushing the headers a bit
* Make custom::task_arena inherited from tbb::task_arena
* change to the latest TBB API, and more debug printfs
* code-style
* ARM compilation
* aligned "failed system config" between OV and TBB (by using '-1')
* macos compilation fix
* default arena creation (to make sure all code-path have that fallback)
* Incapsulate all TBB versions related logic inside the custom namespace
* Move custom layer header to internal scope + minor improvements
* with all NUMA/Hybrid checks now consolidated in the custom_arena, cleaning the ugly ifdefs thta we had
* Introduce new ThreadBindingType + fix compilation
* fixing OMP compilation
* OpenVINO Hybrid CPUs support
* Remove custom::task_arena abstraction layout
* Get back to the custom::task_arena interface
* Add windows.h inclusion
* Fix typo in macro name
* Separate TBB and TBBbind packages
* Fix compile-time conditions
* Fix preprocessors conditions
* Fix typo
* Fix linking
* make linking private
* Fix typo
* Fix target_compile_definitions syntax
* Implement CMake install logic, update sha hash for the tbbbind_2_4 package
* Add tbbbind_2_4 required paths to setup_vars
* Update CI paths
* Include ie_parallel.hpp to ie_system_conf.cpp
* Try to update dependencies scripts
* Try to fix dependencies.bat
* Modify dependencies script
* Use static tbbbind_2_4 library
* Remove redundant paths from CI
* Update tbbbind package version
* Make custom::task_arena inherited from tbb::task_arena
* Incapsulate all TBB versions related logic inside the custom namespace
* Move custom layer header to internal scope + minor improvements
* Introduce new ThreadBindingType + fix compilation
* Fix compilation
* Use public tbbbind_2_4 package
* fixed macos build, corrected comments/desc
* reverted to the default binding selection logic ( to preserve the legacy beh)
* OpenVINO Hybrid CPUs support
* Remove custom::task_arena abstraction layout
* Get back to the custom::task_arena interface
* Add windows.h inclusion
* Fix typo in macro name
* Separate TBB and TBBbind packages
* Fix compile-time conditions
* Fix preprocessors conditions
* Fix typo
* Fix linking
* make linking private
* Fix typo
* Fix target_compile_definitions syntax
* Implement CMake install logic, update sha hash for the tbbbind_2_4 package
* Add tbbbind_2_4 required paths to setup_vars
* Update CI paths
* Include ie_parallel.hpp to ie_system_conf.cpp
* Try to update dependencies scripts
* Try to fix dependencies.bat
* Modify dependencies script
* Use static tbbbind_2_4 library
* Remove redundant paths from CI
* Update tbbbind package version
* Make custom::task_arena inherited from tbb::task_arena
* Incapsulate all TBB versions related logic inside the custom namespace
* Move custom layer header to internal scope + minor improvements
* Introduce new ThreadBindingType + fix compilation
* Fix compilation
* Use public tbbbind_2_4 package
* Apply review comments
* Fix compilation without tbbbind_2_4
* Fix compilation with different TBB versions
* code review remarks
* fix for the NONE pinning code-path under HYBRID_AWAR
* whitespace and cleaning the debug printfs (per review)
* code-review comments
* fixed code-style
Co-authored-by: Kochin, Ivan <ivan.kochin@intel.com>
Co-authored-by: Kochin Ivan <kochin.ivan@intel.com>
