Jan Iwaszkiewicz
17914d516a
[ONNX] QLinearMatMul ( #7418 )
...
* Add qlinear_matmul files
* Extend matmul helpers
* register op in bridge
* Add test for 2d
* newlines at end of files
* Codestyle changes
* Re-use of nodes inputs
* Remove xfails
* Register as set1
* reshape inputs as scalars
* remove fixed xfail issue residues
* update changes to fit the newest master
Co-authored-by: dkozykowski <dawid.kozykowski@intel.com>
2021-10-20 11:02:14 +03:00
Gleb Kazantaev
88f56669e2
Remove Provenance tags ( #8101 )
2021-10-20 10:28:58 +03:00
Ilya Lavrenov
4f203b9a2e
Fixed ENABLE_NCC_STYLE condition ( #8092 )
2021-10-20 09:01:59 +03:00
Kelvin Choi
c454e414bb
[GPU] Add unittests for fusing reorder bfyx to fsv32 ( #8080 )
...
Signed-off-by: Kelvin Choi <kelvin.choi@intel.com>
2021-10-20 13:57:18 +09:00
Mingyu Kim
a1c77879d5
[GPU] Select impl_type after layout propagation ( #8054 )
...
pooling layout is not chosen initially in reorder_inputs.
In that case, cldnn pooling is chosen instead of onednn.
2021-10-20 12:19:46 +09:00
hyunback kim
6a63cc1936
[GPU] Apply fusing/reshape changes into master. ( #8002 )
...
* [GPU] Apply fusing/reshape code into master.
Merge prepare_primitive_fusing, handle_reshape into master.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* Apply code-review.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* Apply code-review
Signed-off-by: hyunback <hyunback.kim@intel.com>
2021-10-20 09:08:39 +09:00
Roman Lyamin
03424849fb
[GPU] Add separate config for each device ( #7421 )
2021-10-19 16:51:38 +03:00
Xiping Yan
b7ccdf4490
Xp/fix jira 6671 atanh fail mac wins ( #8058 )
...
* tmp
* Fix atanh correctness test issue and veriyf pass.
Signed-off-by: Yan, Xiping <xiping.yan@intel.com>
* Only keep available input value for correctness test.
Signed-off-by: Yan, Xiping <xiping.yan@intel.com>
2021-10-19 14:33:49 +03:00
Ilya Churaev
4964ec890c
Add add_output API for ov Function ( #7980 )
...
* Add add_output API for ov Function
* Fixed comments
2021-10-19 14:07:48 +03:00
Maxim Andronov
7ee322be90
[CPU] Eltwise node update and tests ( #7892 )
...
* [CPU] Eltwise node update and tests
* comments applied
* WA for setShape for TensorDesc with blocked layout
2021-10-19 13:33:15 +03:00
Anastasia Kuporosova
bfb092a6d6
[Python API] Remove pyngraph mark in ie python tests ( #8013 )
...
* [Python API] Remove pyngraph mark
* move ngraph import from test functions
2021-10-19 13:14:12 +03:00
Dmitry Pigasin
e3d5d4520f
[IE Sample Scripts] Change omz tool calls ( #7717 )
...
* Change omz tool calls in sample scripts (linux)
* Change omz tool calls in sample scripts (windows)
* Remove venv creation and /tools/model_optimizer/requirements.txt installation
* Remove `open_model_zoo/tools/downloader` from setupvars
* Remove `accuracy_checker` from setupvars
* Add error if omz tools are not installed
* Update scripts/demo/run_sample_benchmark_app.bat
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
* Add check for all omz tools
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
2021-10-19 11:49:42 +03:00
Krzysztof Bruniecki
feb7814a96
[GNA] Fix pooling output backward compatibility ( #7889 ) ( #7966 )
...
* [GNA] Fix pooling output backward compatibility
* Apply review
2021-10-19 11:43:59 +03:00
Mateusz Tabaka
bcc1404c58
[ONNX] remove unnecessary nodes when handling Identity op ( #8060 )
2021-10-19 10:18:13 +02:00
Vladislav Golubev
c84db94697
[dynamism][CPU] MatMul transformations: dynamic shapes support ( #7582 )
...
* [TESTS] CPUUnitTests: unitTestUtils added to CMakeLists
* Reshape1D: functions moved to namespace
* [nGraph] PartialShape extending
* CPU specific transsformations for FullyConnected enabled to work with dynamic shapes
* [Transformations] FC transformations: dynamic shapes support
* [Transformations] FCBiasFusion: removed legacy check on bcast on weights
* [TESTS] FC transformations: tests
* [Transformations] SmartReshape: avoid tranformation utils methods
* codestyle fix
* codestyle fix
* [TESTS] MatMulTransformation tests: compilation error fix
* [CPU] FullyConnected: shape inference
* postreview fixes
* [CPU] FullyConnected: shape inference fix
* [nGraph] PShape dimensions insertion fixed
Co-authored-by: Stepyreva, Evgenya <evgenya.stepyreva@intel.com>
2021-10-19 10:59:29 +03:00
Mateusz Tabaka
18aaaa79a0
Disable StridesOptimization for shapes with dynamic spatial dimensions ( #8065 )
2021-10-19 10:15:17 +03:00
Luo Cheng
602fb74fe4
[PDPD Frontend] enable bicubic, trilinear, linear ( #7731 )
2021-10-19 08:48:26 +03:00
hyunback kim
88cab67833
[GPU] Add fused operation test for concat and pooling ( #7900 )
...
* [GPU] Add fused operation test for concat.
And fused test and enable post ops in concat onednn integration.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* Add depence code when machine does not support imad.
oneDNN post operation has issue.
Signed-off-by: hyunback <hyunback.kim@intel.com>
2021-10-19 14:02:08 +09:00
Anton Pankratv
3f8856862d
Forbid typed data access and roi tensor creation for little bit types ( #8007 )
2021-10-19 04:53:33 +03:00
Ilya Lavrenov
00be1d5a15
Fixed byteSize for Blob / ov::Tensor ( #7989 )
2021-10-19 02:01:10 +03:00
Jan Iwaszkiewicz
78358bacb0
[PYTHON API] Expose ngraph bindings as part of openvino API ( #8027 )
2021-10-18 21:36:48 +02:00
Ivan Novoselov
cf52ba5c08
[CPU] Bug/custom_inplace_reorders/fix unit ( #7222 )
2021-10-18 18:31:23 +03:00
Mang Guo
ac9c2f19dc
[CPU] Register pass to remove unnecessary convert layers ( #7642 )
2021-10-18 15:28:47 +03:00
Steve Yoo
4c722e025b
Migrate ngraph backend test of condition operations ( #7907 )
...
* Add test cases for PReLU in cpu plugin
* For case when slope is vector
* Add Bucketize and NonZero template plugin reference tests
* Apply supported types for NonZero
* Edit float_t and double_t to float and double
* Edit std::vector<bool> to std::vector<char>
2021-10-18 13:20:09 +03:00
Ilya Lavrenov
62e631a2b2
Enabled HETERO:CPU with dynamic shapes ( #8049 )
2021-10-18 11:04:49 +03:00
Mingyu Kim
99544d2a43
[GPU] Improve opencl compile error message ( #8032 )
...
Show some details on error message by parsing "error" in the message.
2021-10-18 15:44:46 +09:00
Mingyu Kim
ae2913d3b5
Duplicated: OneDNN`s Convolutions and Deconvolutions support ( #8029 )
...
* [GPU] Fix incorrect reusage of OneDNN postops configurations
* [GPU] Add OneDNN's convolutions and deconvolutions support
* [GPU] Do not run fusing unittest when imad is not supported
Co-authored-by: Sergey, Shlyapnikov <sergey.shlyapnikov@intel.com>
2021-10-16 17:24:44 +09:00
Ilya Lavrenov
b2977632d7
Fixed copy-paste in docs ( #8046 )
2021-10-15 19:53:16 +03:00
Pavel Esir
f661133242
[MO] fixed dealing with None in then/else branches of Select ( #7465 )
...
* fixed dealing with None values in then/else branches of Select
* generalized solution when condition is non one elemental
* fix when both branches are None
* minor corrections
* rewritten Select unit-tests; fixed for condition with [True] mask
* removed mutable default arg from build_graph, added a few more test cases with masked condition, other minor corrections
* corrected output_shape calculation when broadcasting is off
* layer tests fixed: relaxed assert for condition shape to let pass TF Select
* corrected shape calculation when condition is not elementwise equal, calculated calculation for TF, corrected calculation of shape when values are not set
* fixed a typo for Select from TF Where
2021-10-15 18:51:52 +03:00
Egor Duplensky
5b075d8575
[CPU] Enable back INT32 for MatMul ( #7898 )
...
- Also enabled ngraph onnx matmul integer tests
2021-10-15 18:05:23 +03:00
Alexander Kozlov
bbeec714aa
Moved Post-training Optimization Tool to open-source ( #7940 )
...
* Moved POT to opensource
* Added OMZ as a submodule
* Exclude OMZ from ShellCheck
2021-10-15 16:35:35 +03:00
Polina Brzezinskaya
2e4514b4df
[IE][VPU] Fix scaling preprocessing in myriad plugin ( #7769 )
...
This pr fixes scaling preprocessing implementation in MYRIAD plugin
2021-10-15 16:18:56 +03:00
Ilya Znamenskiy
2079563e4d
[GPU] Deleted OpenCL max WGS limitation ( #7107 )
2021-10-15 15:06:16 +03:00
Anton Chetverikov
e034a072ea
[MO] Fix NMS 3rd output shape ( #7992 )
...
* Fix NMS 3rd output shape
* Add tests for NMS_5 shape infer
* Add comments, fix codestyle
2021-10-15 14:44:08 +03:00
Irina Efode
e8f2249d8e
Fix master branch and resolve conflicted PRS ( #8039 )
2021-10-15 14:05:39 +03:00
Mateusz Bencer
ccdd0e61d5
Fixed expanding ONNX function ( #7747 )
...
* fix expand_onnx_functions
* refactor + unit test
* fixed function in function case
* fixed expand_onnx_functions
* fixed default value of shape in ValueInfo
* enable xpass model
* changed MergeFrom to Swap
* added xfail with missing test data
* added more unit tests
* styles applied
* used std::rotate, review remarks
* removed debug code
* after offline discussion remarks
* fix checking input/output names on Windows
* names comparator refactor
* replace regex with custom comparison
* review remarks
2021-10-15 13:40:28 +03:00
Katarzyna Mitrus
96df1a14ce
Slice-8 Reference implementation ( #7749 )
...
* Init template reference tests
* Move output shape calculation to separate function
* Add to evaluate_map (reuse old ref)
* Add negative step test
* Make slice evaluate template type independent
* Enable typed tests for reference
* Add more reference tests
* Revert old slice ref changes
* New Slice reference implementation POC
* N-dim generalization
* Code refactor
* Style alignment
* Add more checks
* Add more tests
* Add zero dims/empty output tests
* Simplify typed tests generation
* Slice op leftovers
* Add unsigned types reference tests
2021-10-15 12:41:39 +03:00
Vladimir Paramuzov
a091201b67
[GPU] transformations pipeline is extracted to separate class ( #7954 )
2021-10-15 12:39:57 +03:00
Eugeny Volosenkov
3e48008c3f
Unroll If in ngraph ( #7403 )
...
* added unroll_if
* Fix interface function
* Fix code style
* add false test for unroll if
* add if transformation in subgraphs
* fix code style
* fix transformations
* delete internal transformations
* Fix comments
* fix to replace_node
* fix description fo transformation
* fix CompileNetwork
* fix comments
* fix comments
* add function get_ie_output_name(input);
* fix code style
* disable cpu test
2021-10-15 12:37:43 +03:00
Irina Efode
3f690314fa
[IE TESTS] Behavior2.0 -> Plugin ( #7579 )
...
* Need to remove
* Preprocessing + InferRequest
* infer_request_callback
* tmp commit
* InferRequest
* tst
* tetet
* Fix template
* SetBlobType
* InferRequest Conformance
* Add prc
* Remove prc
* Fix paths
* commit
* Fixes for review
* Dynamic Batch
* fix
* fiz
* Try
* fkf
* Fix azure
* Centos
* skip
* conformance fix
* fix template
* Small fix
* small fix in print
* Fixes
* Behavior 2.0 -> ExecutableNetwork
* Export
* Skip configs
* Incorrect config
* Extend ExecNet
* add_out
* GnA+CPU
* fix template
* fix centos
* Init
* init
* Fix merge
* Separate OV_CORE_INTEGRATION
* vers
* plugin
* Remove test_plugin
* plugin
* test
* configuration
* config
* core_integration
* Small refactor
* Clean up exec net
* Config
* Clean&Up behavior tests
* Remove extra
* query network
* Version
* Add new tests
* merge
* fix win
* Fix some ci problems
* Refactor
* gpu
* myriad
* template
* gna
* Fix all plugin
* ov2.0 exec_nry
* openvino2.0
* build
* fix ci
* fix ci
* fix myriad
* Try to fix ci
* Try
* try
* Update windows.yml
* Update windows.yml
* Revert "Update windows.yml"
This reverts commit 93e326229c5eb4bf875a351ede4e1e
2021-10-15 11:47:50 +03:00
Wilson Seok
34fce43ab0
Migrate ngraph backend test/convolution backprop ( #7997 )
...
* Remove fp16 of Convert layer test from skip_tests.config.cpp as it works now
* update repo
* add draft reference of binary convolution
* build test
* fix build error
* add binary_convolution reference test and ngraph visitor api test
* remove ngraph backend test of binary_convolution
* add reference test and remove ngraph backend test of convolution_backprop
* add f64 case for convolution_backprop
2021-10-15 11:39:04 +03:00
Evgeny Kotov
a9688bac54
Broadcast const pass ( #7746 )
...
* restore from another devel branch
* cleanup
* cleanup
* code review fixes
* code review fixes
* use Broadcast instead of referenct
* unit tests reorginize
* add comment
* cleanups
* determine broadcast type
* unit test using Bool instead of {true, false}
* broadcast types fixes
* move SubstituteScaleShiftBroadCastPass under if (!isNgraphPassesUsed)
* windows build fix
2021-10-15 11:27:36 +03:00
Anton Pankratv
4cc53c97a1
Blob processing in HETERO forwarded to sub requests ( #8012 )
...
* Blob processing in HETERO forwarded to subrequests
* Added template dynamic hetero tests
2021-10-15 10:44:11 +03:00
Sergey Shlyapnikov
fa38103e5b
[GPU] Add additional rule for oneDNN eltwise memory reuse ( #7924 )
2021-10-15 10:42:00 +03:00
Mikhail Ryzhov
a4cc31c0b9
Shared memory_solver class ( #7918 )
...
* Shared memory_solver class
* fixed clang format
2021-10-15 09:58:55 +03:00
Sergey Shlyapnikov
345033f3e1
[GPU] Keep -Os optimize option only for Release build ( #7922 )
2021-10-15 09:56:18 +03:00
Sergey Shlyapnikov
8ea9986896
[GPU] Add additional handling of fused activations for OneDNN primitives ( #8004 )
2021-10-15 09:31:49 +03:00
Anton Chetverikov
897cd09a5a
[MO] IR Reader updates for support IR generated by PDPD FE ( #7646 )
...
* Add extender for LSTMSequence operation
* Update BOM file
* Update gather shape inference function to support 1D tensors as axis
* Refactor split_normalize_output function usage
* Update temporary Result operations processing
* Revert fake split output renaming
* Revert kaldi assign result removing
* Add assert for node type
* Update model-optimizer/mo/utils/ir_reader/layer_to_class.py
Fix comment
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
* Update model-optimizer/extensions/ops/gather.py
Update axis value check.
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
Co-authored-by: Evgenya Stepyreva <evgenya.stepyreva@intel.com>
2021-10-15 09:29:07 +03:00
Mingyu Kim
a25668d106
[GPU] dryrun bugfix ( #7998 )
...
dryrun failed in case nested graph is created.
(due to constant propagation)
For nested program, do not skip compile() to make outer program compiles
OK.
2021-10-15 15:28:48 +09:00
Vladimir Paramuzov
55792e9a00
[GPU] Enable sub-device selection during shared context creation ( #7976 )
...
* [GPU] Enable sub-device selection during shared context creation
* [GPU] Fixed code style
* [GPU] Added default value description for target tile id
Co-authored-by: Mikhail Letavin <mikhail.letavin@intel.com>
2021-10-15 09:20:09 +03:00
