* Updated conversion of TF OD API 2.4 SSD models
* Fixed issue when more Conv2D nodes were selected for weights permutation when converting TF OD API models
* Code style fixes
* Fixed code comments
* bulk change `type T` to `type *T*` in spec
* update all `T` which referee to type to use *T* pattern
* get back with `T` where T is dimension
* fix *T*1 -> *T1*
* Make italic types where was no formating
* add priors to loader and counts transformation
* fixes in select insertion for case with context gathering - LSTM - context gathering
fix for edge parallel to ReadValue
extend counts option to case of priors inside mdl model file
* fixed tests
* fixed typo
* fixed issue with input names
* fix priors loading + comments
* fix e2e test: error with not found transformation
* print debug info for dependency graph - should be reverted
* should be reverted: debug commit
* Revert "fix e2e test: error with not found transformation"
This reverts commit 8320fa99bf.
* revert debug commits
* fixes after review
* review fixes
* review change
* review changes
* Added pugixml as submodule
* CVS-34900: updated pugixml to v1.11.4
* Fixed link with pugixml
* USe pugixml::static
* Try to fix bug
* Removed GITHUB_PULL_REQUEST
* Replaced OpenVINO_MAIN_SOURCE_DIR -> OpenVINO_SOURCE_DIR
* Removed some usages of IE_MAIN_SOURCE_DIR
* Use ngraph target directly
* Fixed return in `run_timetest`
* Fixed `parse_stats()` argument type notation
* Pass message from `timetest` exe to test error msg to be uploaded to DB
* Written header file for the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Small fixes.
* Started to write cpp-file for the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Written the postprocessing function for the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Code style fixes.
* Started to write calculations of the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Written the helper function refine_anchors().
* Written the helper function unpack_boxes().
* Small fix.
* Code style fixes.
* Written calculations of the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Code style fixes.
* Fixed type of one argument of fill_output_blobs().
* Code style fix.
* Written evaluate() for the operation ExperimentalDetectronGenerateProposalsSingleImage (in evaluates_map.cpp).
* Deleted redundant 'using' for typedef.
* Continued to write tests for ExperimentalDetectronGenerateProposalsSingleImage evaluation in the nGraph reference implementation.
* Fixed the reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Deleted commented code. Code style fixes.
* Code style fix.
* Deleted debug prints in the test for an evaluation of the nGraph operation ExperimentalDetectronGenerateProposalsSingleImage.
* Enabled ONNX Importer test for the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Enabled some test.
* Small fix.
* Now postprocessing of the nGraph reference realization of the operation ExperimentalDetectronGenerateProposalsSingleImage throws exception when input data type is not supported.
* Code style fix.
* Written yet another test for the nGraph reference implementation of the operation ExperimentalDetectronGenerateProposalsSingleImage.
* Reused ProposalBox type from the reference implementation of the operation Proposal.
* Added check for the attribute post_nms_count in the operation ExperimentalDetectronGenerateProposalsSingleImage evaluation.
* Small fixes.
* [MO] Fix showing graceful error message when no ngraph case
Move frontend-related imports inside appropriate functions
And use try/except if import fails to show graceful error message
* Revert "[MO] Fix showing graceful error message when no ngraph case"
This reverts commit 72d52bd946.
* Print error before running of subprocess if IE or nGraph Python API is not found
* Added pytest to requirements_dev.txt
* Fixed review comment
* Added compile_tool to PATH
* Installed benchmark_app
* Fixed ngraph unit tests in Debug
* Install of VPU_FIRMWARE to <root>/tests folder
* Conditional installation of benchmark_app
* Update docs
* create type_prop test for cos
* Create visitor test foor cos
* add RTTI macro
* Create separate ref impl for integer input
* add cos to SLT integer activation type
* add backend test for int input
* add cos to constants.py
* style apply
* fix spec
* add ngraph check for outputs and inputs
* fix inputs ngraph check
* remove unnecessary includes
* style dix
* Use ngraph rtti macros
* add visitor test
* add visitor test cases to cover all possibilities
* add boolean case to type prop tests
* add boolean type to sslt
* add nonzero to trusted op list
* multiply dimensions when input shape is dynamic
* Add and disable cpu instantiations of slts
* Add missing coma in skip tests config for cpu
* Fix expected output shapes in type prop testcases
* add partialshape input cases to type prop tests
* Use pool stride instead of window size where applicable
* Add test for pooling stride not equal to wnd
* Add more tests and cleanup
* Fix SW_FP32 legacy cnn
* [WIP] Refactor CNN1D
* Remove unused (commented out) code
* Add tests
* DRAFT: hot-swap async request (when VPU or other acceleratore finally loaded)
* DRAFT2: communicating the hot-swap back to the AutoExecNetwork, added logic for blobs (and keeping them internally to the AutoRequest), added lot's of comments for fixme (items to close the semantics and behaviour)
* Rebase/refactor code and fix tests issue
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Fix logic for AutoExecutableNetwork
1. force loadNetwork in parallel by std::launch::async
2. address some fixme in auto_exec_network.cpp
3. capture by reference
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Fix core dump
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Lambda explicit capture by value
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Add debug log to detect destory order of plugin and core
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Use sync to load cpu and gpu to check ci
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Revert "Use sync to load cpu and gpu to check ci"
This reverts commit 66e09ccd47321e26f68392976d59b1e69cd3df1a.
* Copy network in lambda
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Try to fix CanCreateInferRequest test in Centos
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Remove print log
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Fix CI issues in AUTO because GPU execNetwork doesn't support SetConfig
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* Address reviewer's comment: handle load network failure
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
* weak_ptr rather than plain ICore* to make sure we hold the Core (whihc holds the plugins in turn) from the destruction in case we need that
* Replace ie::ICore* to shared_ptr<ie::ICore>
Signed-off-by: Shoujiang Ma <shoujiang.ma@intel.com>
Co-authored-by: myshevts <maxim.y.shevtsov@intel.com>
* fixed multi setConfig to throw on unsupported config keys
* fixed hetero setConfig to throw on unsupported config keys. simplified the test on the incorrect keys
* aligning MULTI tests to the new (fixed) behaviour - also aligned with AUTO and HETERO
* fixed multi setConfig to throw on unsupported config keys- updating the MYX tests
