Anastasia Kuporosova
c7eeda0247
[Python API] Align openvino/__init__ files ( #9871 )
...
* [Python API] Align openvino/__init__ files
* style fix
2022-01-26 21:16:34 +03:00
Mikhail Letavin
a90bb1ff01
[GPU] Force shared surface cache cleanup in RemoteContext ( #9816 )
2022-01-26 19:19:29 +03:00
Nikolay Tyukaev
e0c10c1a6f
nbdoc cmake ( #9869 )
2022-01-26 18:45:16 +03:00
Zhang Yi
f23714fe65
fix klocwork scan ( #9785 )
2022-01-26 18:26:55 +03:00
Mikhail Nosov
6d40e5df96
'If' deserialization fixes ( #9881 )
2022-01-26 17:50:43 +03:00
Egor Shulman
68423cef40
[CPU] Coverity fixes ( #9838 )
2022-01-26 17:43:28 +03:00
Xin Wang
26ccbe9ebc
Change internal function name to avoid loop issue. ( #9890 )
...
* Change internal function name to avoid loop issue.
* Fix comment.
2022-01-26 17:35:31 +03:00
Anastasia Popova
84f2e9fc24
Parameter/Result tensor names fix ( #9754 )
...
* Added tensor names checking transformation. Added framework name saving to tensor names list for input/output cut.
* Small fixes.
* Fixed tensor propagation in PowerToEltwises.
* Corrected tensor checking.
* Corrected tensor checking.
* Fixed MemoryOffsetAdjustment().
* Fixed tensor propagation for Yolo, ONNXMaskRCNNTransformation().
* Small fix.
* DetectionOutput tensor name set.
* Tensor name set for Reshape node in OD API.
* Temporarily added set of tensor names in ConvertGroupedStridedSlice.
* Small corrections, added tests.
* Added checks.
* Added deafault names setting.
* Moved default names setting to single place.
* Added port normilize befor setting tensor names.
* Fixed ResultRename logic.
* Removed tensor setting of unset ports.
* Corrected input cut tensor naming.
* Corrected InputCut, renamed set_tensor_names()->add_tensor_names().
* Fixed tensor setting for InputCut.
* Code corrections.
2022-01-26 17:23:46 +03:00
Mikhail Nosov
8f94d6dd3f
Fix unity build for 'inference_engine_s' target ( #9899 )
...
Without fix, build of 'cpuUnitTests' will fail as 'inference_engine_s' will contain:
IE_STATIC_DEPENDENT_FILES = file_utils.cpp
$<TARGET_OBJECTS:${TARGET_NAME}_obj> - containing 'unity_cxx' which includes 'file_utils.cpp'
This causes multiple definition error of all methods inside file_utils.cpp
2022-01-26 17:11:40 +03:00
Alexandra Sidorova
1d82294e00
[CPU] Fixed dynamic Concat with negative axis ( #9861 )
2022-01-26 17:10:34 +03:00
Vladislav Volkov
c026f8348b
[CPU] Fix for memory leak in MKLDNNTensorIteratorNode ( #9903 )
2022-01-26 16:59:41 +03:00
Anton Dudchenko
022abf8614
[IE][VPU]: Don't compile myriad custom layers and tools for ARM ( #9912 )
2022-01-26 16:30:32 +03:00
Maksim Doronin
e46b629950
[IE][VPU]: dynamic hswish and gather downgrade ( #9915 )
...
* Dynamic HSwish
* Gather8 to Gather7 downgrade
2022-01-26 16:28:03 +03:00
Mikhail Nosov
2ea52bae6e
Fixes of plugins.xml generation ( #9922 )
...
- Added registration (and unregistration) of ov_auto_batch_plugin. Otherwise 'BATCH' plugin will always produce new XML line without removing old one
- Added unregistration of legacy plugin names (<= 2021.4 release). Otherwise old lines like "libHeteroPlugin.so" will not be removed from plugins.xml file
2022-01-26 16:00:04 +03:00
Indira Salyahova
17c394c8fe
[POT] Update patterns ( #9786 )
...
* Update special_patterns.py
* add pattern in docs
2022-01-26 14:59:51 +03:00
Indira Salyahova
cf49cd6ce8
fix: check batch bias shape equal one in fbc ( #9907 )
2022-01-26 14:37:43 +03:00
Ilya Churaev
251897a91a
Removed old cmake scripts ( #9876 )
2022-01-26 14:24:30 +03:00
Ilya Churaev
d2c92d26be
Added offline transformation to all target ( #9893 )
2022-01-26 14:23:59 +03:00
Irina Efode
ed273ab47a
[IE TESTS][IE CONFORMANCE] Allow to specify body extracting in statistic and subgraphDumper ( #9797 )
...
* [IE TESTS][IE CONFORMANCE] Allow to specify body extracting in statistic and subgraphDumper
* Fix link
2022-01-26 11:35:36 +03:00
Nikita Malinin
1524ff898a
Update IEEngine with annotation for async ( #9909 )
2022-01-26 11:13:01 +03:00
Mingyu Kim
1d417e888b
Spacing fix ( #9894 )
...
* Update benchmark_app.hpp
* Update benchmark_app.hpp
2022-01-26 08:46:05 +03:00
Milana Shhanukova
d81d7f15a1
do not change shape for scalars ( #9751 )
2022-01-25 19:37:34 +03:00
song, bell
f2e5788035
use priority queue for timeout issue ( #9765 )
...
* use priority queue for timeout issue
Signed-off-by: fishbell <bell.song@intel.com>
wrap tbb pq
Signed-off-by: fishbell <bell.song@intel.com>
clang format
Signed-off-by: fishbell <bell.song@intel.com>
fix potential dead-lock
Signed-off-by: fishbell <bell.song@intel.com>
fix arm build
Signed-off-by: fishbell <bell.song@intel.com>
try fix build on arm
Signed-off-by: fishbell <bell.song@intel.com>
merge master
Signed-off-by: fishbell <bell.song@intel.com>
* remove debug info
Signed-off-by: fishbell <bell.song@intel.com>
* use atomic to replace the lock
Signed-off-by: fishbell <bell.song@intel.com>
* clang format
Signed-off-by: fishbell <bell.song@intel.com>
2022-01-25 18:44:46 +03:00
Eugeny Volosenkov
b57f1b0fb3
Delete check on port id in InputCut ( #9796 )
...
* delete refer_to_faq_msg(31)
* fix documentation
* fix documentation
2022-01-25 17:41:08 +03:00
Indira Salyahova
0ecbe50bf5
[POT] Optimize quantization scheme for SPR/ATS ( #9564 )
...
* feat: delete fq from branch
* delete transformation
* pylint
* delete gpu_ats
* add tests
* pylint
2022-01-25 17:37:45 +03:00
Nikita Malinin
6927547f9c
[POT] Update IEEngine with raw output processing ( #9777 )
...
* Update IEEngine with raw output processing
* Update ieengine output layers naming
* Pylint fixes
* Update replace with the regex method
* utils.py newline
* Pylint fix
2022-01-25 17:37:12 +03:00
Yury Gaydaychuk
365bd7c46e
[CPU] DefConv fixed for small channels ( #9718 )
2022-01-25 17:26:00 +03:00
Vitaliy Urusovskij
0282f11165
Fix control flow issue ( #9865 )
...
* Fix control flow issue
* Extend exception messages in TF PB parsing
2022-01-25 17:23:59 +03:00
Eugeny Volosenkov
439e3b26aa
Model Optimizer incorrectly convert per-channel Quantize/Dequantize operatros to FakeQuantize ( #8321 )
...
* add q_dq_resolver
* prepare FQ for offline transfornations
* fix quantize_dequantize_linear_resolver.py
* fix tools/mo/openvino/tools/mo/middle/dequantize_linear_resolver.py
* enable compress_quantized_weights.py
* fix sub case
* delete force_clean_up
* add isolated attribute
* add test for quantize_dequantize_linear_resolver
* fix comments
* apply comments
* aplying comments
* applying comments
* applying comments
* Replace q_dq_resolver using pattern
* clean imports
* fix imports
* fix pattern
* delete refernce to unit tests
2022-01-25 15:35:43 +03:00
Pavel Durandin
9f343fa37f
Device memory fix for benchmark app ( #9880 )
2022-01-25 15:10:34 +03:00
Mateusz Tabaka
fa605a2760
Fix begin and end mask in StridedSlice in builder::opset1::flatten ( #9837 )
2022-01-25 13:04:36 +01:00
Maxim Andronov
94d403a1eb
[CPU] Don't set subnormals to zero for rnn packed format ( #9836 )
2022-01-25 14:58:41 +03:00
Sergey Lyubimtsev
54203a2095
Fix for setting PATH to openvino libs installed from pip in windows conda env ( #9892 )
2022-01-25 14:58:19 +03:00
Anton Chetverikov
88903ee7ae
Remove incorrect layout changing ( #9764 )
2022-01-25 14:10:06 +03:00
Mang Guo
9a522137bf
Hot fix fir regression on Mac ( #9895 )
2022-01-25 13:47:37 +03:00
Ekaterina Aidova
2385a16e80
[OMZ]: try update submodule ( #9830 )
...
* [OMZ]: try update submodule
* check conflict resolving
2022-01-25 13:29:30 +03:00
Jan Iwaszkiewicz
1e58c55678
[PYTHON] Allow ports as inputs for infer methods ( #9839 )
...
* Use of ports for input values
* Refactor tests
* Fix inputs overwritting
* Fixed for review comments
* Remove unused branch and refactor ie_api.py
2022-01-25 13:19:12 +03:00
Anton Pankratov
f45991bd64
OV 2.0 c++ configuration API ( #9870 )
...
* New configuration API
* Review fix
* review coments
* fixed device name map
* fixed header
* code stile
* fixed optimization capabilities
* flatten properties
* dox fix
* doc
* merge conflicts
* fixed merge conflicts
* Fixed subobject linkage warning
* foramt fix
* Fixed unity test build
* Merge conflict fixes
* Fixed variadic
2022-01-25 12:43:56 +03:00
Evgenya Stepyreva
3bd6eb6dcd
Dimension tracking: Label propagation along with partial value propagation ( #9834 )
2022-01-25 12:35:56 +03:00
Evgenya Stepyreva
4a85982427
mark sensitive inputs in all Interpolate ctors ( #9883 )
2022-01-25 12:04:27 +03:00
Victor Kuznetsov
013d3f01a0
Change thresholds for MemCheck nightly ( #9879 )
...
* add performance hint to time infer
* change th-ds for mtcnn-o
2022-01-25 11:30:51 +03:00
song, bell
2b9e726040
enable batching plugin through auto ( #9520 )
...
* initial commit
Signed-off-by: fishbell <bell.song@intel.com>
* remove debug code
Signed-off-by: fishbell <bell.song@intel.com>
* fix bug with CPU,GPU
Signed-off-by: fishbell <bell.song@intel.com>
* refine code and test case for auto batching
Signed-off-by: fishbell <bell.song@intel.com>
* optimize the batching query
Signed-off-by: fishbell <bell.song@intel.com>
* refine test case
Signed-off-by: fishbell <bell.song@intel.com>
* optimize if/else
Signed-off-by: fishbell <bell.song@intel.com>
* fix sporadic crash
Signed-off-by: fishbell <bell.song@intel.com>
* refine logic
Signed-off-by: fishbell <bell.song@intel.com>
* refine case
Signed-off-by: fishbell <bell.song@intel.com>
* temp solution to disable batching in auto
Signed-off-by: fishbell <bell.song@intel.com>
* add debug logs
Signed-off-by: fishbell <bell.song@intel.com>
* merge with #9724
Signed-off-by: fishbell <bell.song@intel.com>
* refine the logic
Signed-off-by: fishbell <bell.song@intel.com>
* accurate debug message
Signed-off-by: fishbell <bell.song@intel.com>
2022-01-25 16:28:49 +08:00
Alexey Lebedev
fe5c16ba6a
[tools] Allow to use node name in command line parameters in python benchmark app ( #9874 )
...
* support node names in model specific parameters
* allow to use node name in -i parameter
2022-01-25 11:16:56 +03:00
Pavel Esir
41d4fcdf25
[MO] Add reverse_infer (input rank deducing) for set of operations ( #8928 )
...
* grouped convolution resolver fix
* add reverse_infer for Transpose, Elementwise, Identity and some fixes to successfully OOB convert MXNet models
* add reverse_infer for BatchNorms, dynamic groups fix, multi_box_prior.py fix
* add reverse infer for trivial cases
* input rank setting for image detection ops
* reverse_infer for Gather, AttributedGather add todo comment for Parameter
* reverse infer for LSTM, swapaxis; fixes and unit-tests for Elementwise, Gather, Transpose; also generalized Parameter reverse_infer; fixed DetectionOutput and partial reverse_infer
* clarified comment for LSTM reverse_infer
* detalized Elementwise reverse_infer, added several more unit-tests, Split reverse_infer fix, IteratorGetNext_ext.py fix
* generalized group convolution resolver
* specified in_ports explicitly
* clarified Split, Concat, Parameter, other minor typos
* relaxed compatibility check for elementwise; simplified scalar extracting for convolution.py and gather.py, fixed replacer order for ConvolutionNormalizer.py
* added routine set_input_shapes for image detection ops
* fixed typos to pass locally E2E, slightly changed LSTM reverse_infer
* VariadicSplit reverse_infer bugfix
* IteratorGetNextCut.py added check if port is connected
* explicitly specified in_port, added asserts, improved Split reverse infer
* sorted imports
* reverse_infer for ONNX and MXNet LSTM, GRU, RNN operations
* replaced Error raising with red warning
* added missing return for defined input shapes, fixed getting num groups for tf conv
* added mandatory attributes; fixed failing unit-tests, fixed typo with scalar group in ConvolutionNormalizer
* removed priorbox reverse_infer; Parameter reverse_infer corrected to use the only one existing out_port
* fixed clarify_partial_shape_test
2022-01-25 10:14:41 +03:00
Anastasia Popova
d27bbb4bdd
Preserve inputs and outputs order for TensorFlow 2 and MXNet models ( #9683 )
...
* Preserve outputs order TF.
* Preserve of input/output indices for MxNet.
* Small fix.
* Added check.
* Small fix.
* Corrected Keras model importing.
* Fixed Keras model loading.
* Small correction.
* Corrected model loading.
* Small fix.
* Comment corrected.
* Removed unnecessary import.
2022-01-25 10:13:05 +03:00
Victor Kuznetsov
6f60a7d8f0
First time inference validation for dynamic shapes ( #9671 )
...
* add performance hint to time infer
* add reshape methods for time infer tool
* add func declarations to common_utils.h
* add check if node is static (reshape pipeline)
* changes in desc of reshape funcs
* fix logic operator
* add reshape utils.h | move parsing logic in main func
* add tests_shared_lib to CMakeLists.txt
* change fill blobs with setStaticShapesBlobs
* change delimiter
2022-01-24 16:19:03 +03:00
Anastasia Popova
ce533fc287
Preserve input/output order for ONNX. ( #9352 )
...
* Preserving of input/output indices for ONNX.
* Fixed checks.
* Fixed for case of multiple outputs of node before Result.
* Added test for multiple tensor names before Result.
* Multiple tensor names before Result fix.
* Added order alignment for user input/output.
* Extended for case of input names in Parameter tensor list.
* Fixed unit tests.
* Corrected help.
* Small correction.
* Code refactoring.
* Temporarily reverted refactor.
* Fixed wrong changes.
* Fixed wrong changes.
* Returned reverted refactoring.
* Removed inputs_list from serializing.
2022-01-24 15:50:28 +03:00
Mang Guo
413fee2a86
[CPU] Enable cache for MVN ( #9549 )
2022-01-24 15:07:35 +03:00
tgubanova-lohika
7114253d17
[GPU] Implement Gelu7 operation support ( #9813 )
...
* [GPU] Add Gelu tanh approximation support
* [GPU] Register op7::Gelu operation in gpu plugin.
* [GPU] Add op7::Gelu operation tests to GPU single layer tests
2022-01-24 15:05:34 +03:00
Irina Efode
ec38a88b60
[IE TESTS] Conformance migration to OV2.0 ( #9092 )
...
* Squashed commit of the following:
commit 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
2022-01-24 14:59:15 +03:00
