* Shared preprocessing tests for plugins.
Comparing inference with reference implementation
* Moved evaluate tests to template plugin
* Fixed clang-style
* CPU tests: Set IE precision manually in SetUp. Also allow rounding to integer mismatch
* Added acceptable threshold depending on particular test
* Fixed typo
* Moved ie::Parameter to new API
* New OpenVINO API macro
* Used OpenVINO Plugin API for some plugins
* Used OPENVINO_RUNTIME_API
* Revert "Moved ie::Parameter to new API"
This reverts commit fef5e3b487.
* Fixed code style and docs
* Fixed compilation
* [LPT] INT16, INT32 Quantization support
* [LPT] Support build on platforms with size_t == unsigned int
* [LPT] Test and fix wrong constant
* Fix build for size_t = unsigned int
* [GNA] Add support for DWSC, other fixes and code refactoring.
* [GNA] Change supported layout to NHWC
* [GNA] Detect bias const only on second position, move verification of dwsc to matcher
* Add MulConvFusion transformation
This transformation is applied to a following graph:
```
+-------+ +----------+
| Input | | Constant |
+-------+ +----------+
| |
------ ------
| |
v v
+----------+ +---------+
| Multiply | | Weights |
+----------+ +---------+
| |
----------- ----------
| |
v v
+----------------+
| Convolution Op |
+----------------+
```
and converts it to:
```
+---------+ +----------+
| Weights | | Constant |
+---------+ +----------+
| |
------ ------
| |
v v
+-------+ +----------+
| Input | | Multiply |
+-------+ +----------+
| |
----------- ----------
| |
v v
+----------------+
| Convolution Op |
+----------------+
```
Since 'Weights' are constants in most cases, the right hand side gets constant folded,
and we eliminate Multiply node.
Ticket: 52283
* Handle GroupConvolution, ConvolutionBackpropData, GroupConvolutionBackpropData in separate transformations
* Handle dequantization subgraph
* add namespace
* add more ngraph namespace
* address review comments
* rebasing the perf-modes-2021.3 to the 2021.4
Caveats:
the (explicit) setting #streams is not disabled (as it was before for experiments with DLBenchmark), and the logic slighlty differ (streamsSet)
(cherry picked from commit 1ae1edc0ed)
* overriding streams (to force the TPUT mode to the DLBenchnark)
(cherry picked from commit 7f506cda31)
* disabling reducing #streams to fully mimic baseline c4df94d42d of the 2021.3 (before experiments)
(cherry picked from commit 85073dd1dd)
* clang/identation
(cherry picked from commit 050a4155a9)
* splitting the Transformation to general and CPU specific.
Now hopefully,this fully mimics the baseline c4df94d42d of the 2021.3 (before experiments), as the streams reduce num (as well as early exit on GRU/LSTM/TensorIterator) is deisabled
(cherry picked from commit e98b2c1a67)
* disabling GRU/LSTM/TI + reducing of streams + 5D considered compute-limited only for int8
(cherry picked from commit 32b8d80dee)
* refactored to avoid compute_limited_ratio, reverted the reducing #streams, removed LSTM from limitations
(cherry picked from commit f2b972171b)
* isa-based threshold logic
(cherry picked from commit b218457e1a)
* mode->hint
(cherry picked from commit ec20aa8eca)
* optional PERFORMANCE_HINT_NUM_REQUESTS
(cherry picked from commit 5a3883e3f3)
* moving the perfHints to the common OV config class + initial tests (CPU only, as the actual AUTO/MULTI should be accommodated on the master)
(cherry picked from commit (then fixed)45bafe7d527f466507dea0693aeed51be4ebf776)
* AUTO support for PerfHints
* MULTI support for PerfHints
* Enabling Perf hints for the GPU plugin
* brushing settings output a bit
* disabling "throughput" perf hint being default (until OV 2.0)
* uncommenting the logic which was disabled to force the DLBenchmark to use the throughput mode by default
* removing dead and experimental code, and debug printfs
* clang/code-style
* code-review remarks
* Moved the output of the actual params that the hint produced to the right place
* aligning MULTI's GetConfig beh to HETERO's as captured in the preso (CVS-59960) ratified with the ArchForum
* clang
* benchmark_app brushing
* Update inference-engine/samples/benchmark_app/README.md
* propagating the perf hints thru one more scenario in the merged AUTO-MULTI
* fixed mispint
* Python benchmark_app update for perf hints
* addresssing reviewers comments on the python benchmark_app
* simplifying/brushing logic a bit
* refactor the heuristic to the separate file (to be shared with iGPU soon)
* refactor conversion of modes to the specific GPU config per feedback from Vladimir
* Add visitor api test
* Review ngraph op shell with type_prop tests
* Add op to list of trusted operations
* Change name of struct with information of inputs
* Add include of array data structure to fix windowds compilation error
* Add template plugin test class
* Remove usage of CoordinateTransform index function call from reference implementation
* Rename SLT test suite
* Add template plugin unit test
* Add serialization SLTs
* Add indentation on GatherTreeParams class data members
* [GNA] Fix scale factors propogation for Eltwise with very different inputs ranges
* [GNA] Added test
* [GNA] Added exception for scale factor <= 0
* [GNA] Disable tests with integer weights
* [GNA] Added assert for CNNLayer in getScaleFactor()
* [GNA] Added check if scale factor is inf
* [GNA] Fixed legacy tests
* update comparisiofiles to use const ref param
* introduce conversion layer test definitions
* adapt old tests to the new format
* remove old duplicated conversion tests
* fix "convertion" typo to "conversion"
* fix style issues and abandon unnecessary changes
* fix include order
* update remaining conversion tests to use introduced class
* fix gpu class test naming
* bring back convert.hpp and convert_like.hpp files
* bring back convert.hppcppd convert_like.cpp files
* bring back single_layer_tests/convert.hpp file
* add missing copyright info
* fix issue with braces initiator for conversion types
* add missing convert_like tests
* add deprecated code macros
* update deprecated code macro message
* add missing space in deprecated code macro message
* update skip ConvertLike tests ticket
* update deprecated code to use IE macros
* update remaining ngraph_deprecated macros to use IE macros
* Refactor visitor test of reduce operations to include operation name
* Add template plugin tests for reduction operations
* Remove backent unit tests of reduction operations
* Use Tensor utility to instantiate tests
* Use std::for_each instead of raw for loop to produce test case instances
* Cover remaining backend tests in manifest by SLTs cases
* Fix regex to skip cpu test failing in centos build with precision u64
* Added nGraph as a public dependency
* Fixed Windows warning
* Fixed CMake
* Fixed constant op
* Fixed typo
* Added reshape to PartialShape to CNNNetwork
* Added SetShape to InferRequest
* Enable support of DynamicShape in IE Data
* Add support of dynamic shapes to template plugin
* Fixed build
* Enable support dynamic rank
* Fixed test for dynamic rank
* Fixed some tests
* Fix preprocess tests
* Fixed SetBlob
* Fixed code style
* Add more tests
* Fixed accuracy tests
* Fixed documentation
* Added tests for custom operation
* Added new tests
* WIP: move setShape from infer request to Blob
* Returned isApplicable check back
* Removed obsolete tests for InferRequest::SetShape and add new test for Blob::setShape (a trivial one)
* Fixed artifacts
* Break code style
* Revert "Break code style"
This reverts commit 71ee638cd0.
* Added -j8 for fix_all
* Applied code style fixes
* Adde doxygen items
* Fixed style
* Applied codestyle patch
* Reverted unrelevant commit with template extension
* Fixed cmake file for shared func tests (pick from master)
* Revert all changes in template_extension
* Deleted some old stuff that commented and won't be used
* Fixed wrong exception throwing
* Code style fix
* Fixed preprocessing part
* Fixed incorrect blob reshape in GetBlob
* Deleted incorrect assert in GAPI that prevents passing some tests in Debug
* Fixed issues identified during review
* Removed SetShape, replace getLayoutByDims by getLayoutByRank and removed problematic modification from IE preprocessing
* Fixed comments
* Removed obsolete setShape
* [VPU] Fixed allocating dynamic blobs in myriad_infer_request
* Fixed comments
* Fixed CNNNgraphImpl and comments
Co-authored-by: Ilya Churaev <ilya.churaev@intel.com>
Co-authored-by: Polina <polina.brzezinskaya@intel.com>
