Commit Graph

6728 Commits

Author SHA1 Message Date
Pavel Esir
73e9eb4c61
[MO] add reinterp_shape for StridedSlice (#9622)
* added reinterp_shape for StridedSlice

* package_BOM update

* corrected unit-tests

* returned removed tests
2022-01-31 22:17:15 +03:00
Andrei Molotkov
d5b74b0c6b
[GPU] Fix redefinition variables issue in ExperimentalDetectronROIFeatureExtractor kernel (#9960) 2022-01-31 20:19:16 +03:00
Nadezhda Ageeva
a7910b8869
[GNA] Support new configuration API (#9955)
* [GNA] Support new configuration API

* Review comments: description fixes

* Review comments
2022-01-31 18:48:18 +03:00
Gleb Kazantaev
6dea8d3d86
Update replace_output_update_name method (#10000)
* Update replace_output_update_name method

* Remove wrong legacy test
2022-01-31 18:06:05 +03:00
Lidia Toropova
2f0bcc8542
Moved stress tests to OV API 2.0 (#9791)
* Moved stress tests to OV API 2.0

* Fix for using ouput index instead of get_index

* Removed ov::runtime namespace in stress tests

* Updated stress tests according to latest changes in OV 2.0

* Fix memleaks tests

* Updated run_memcheck.py to process gtest_filter

* Updated fillTensors, added InferAPI1 and InferAPI2 classes

* Updated test_inference_with_streams

* Updated isImage, comments

* Updated fillTensors to fill image_info inputs with positive pseudo-random numbers

* Removed redundant variable in fillTensors
2022-01-31 17:07:25 +03:00
Alexandra Sidorova
252b7d4728
[Benchmark] Fixed preprocessing inputs (#10022) 2022-01-31 17:05:44 +03:00
Anton Dudchenko
89b88af28d
[VPU] Rename INTEL_VPU to INTEL_MYRIAD, move thirdparty and vpu_dependencies (#9827)
* Rename INTEL_VPU to INTEL_MYRIAD, move thirdparty and vpu_dependencies

* Fix thirdparty

* Rename HDDL

* Remove unused files

* Fix HDDL

* Add path to mvnc

* Fix installation on Linux

* Remove thirdparty dir from codeowners

* Move new files

* Remove IE_MAIN_SOURCE_DIR dependencies

Co-authored-by: Maksim Doronin <maksim.doronin@intel.com>
2022-01-31 16:58:33 +03:00
Szymon Irzabek
4eac21e7c0
[GNA] Fix handling reshapes around matmul with 3D/4D Add layer (#9900) 2022-01-31 16:44:35 +03:00
Vladislav Golubev
43130622d3
[Transformations] Convert TI to sequences: dynamic case fix (#9742)
* Convert TI to sequences: dynamic case fix

* tests corrected
2022-01-31 16:37:15 +03:00
Evgenya Stepyreva
cf4d4db4c7
Option not to track dimensions down (#10019) 2022-01-31 16:09:32 +03:00
Anton Pankratov
29d103087a
Aggregating plugins use single config value to pass device list (#9379)
* Aggregating plugins use single config value to pass device list

* format fix

* fixed build

* Fixed tests

* Fixed test

* fied build
2022-01-31 15:43:32 +03:00
Alexey Lebedev
495931673d
[PYTHON API] fix request busy using AsyncInferQueue (#10020)
* Add request.wait() call to get_idle_request_id and wait_all()

* rethrow exception on default callback

* update comment

* fix code style
2022-01-31 15:30:04 +03:00
Egor Shulman
b56fd07169
[CPU] Fixed data storing for AVG Pooling with PostOps on avx2 (#9325) 2022-01-31 14:36:18 +03:00
Indira Salyahova
6145cb736e
[POT] Update patterns with transpose (#9555)
* fix: update pot patterns with transpose

* add tests

* delete tests

* updatereference metrics tf model

* Update test_sanity.py
2022-01-31 14:30:28 +03:00
Mikhail Letavin
70987e327a
[GPU] Remove potentially unsafe references to temporary objects (#9959) 2022-01-31 14:22:59 +03:00
Ilya Sharikov
56f2bc2f12
Add support api 2.0 to layer_tests (#9814)
* Add support api 2.0

* Upd imports

* Update onnx_tests

* Upd tenserflow_tests

* Fix test_split_concat.py

* Fixed comparing
2022-01-31 13:15:30 +03:00
Ilya Churaev
0a2c0e1539
Fixed coverity issues (#10014) 2022-01-31 12:51:52 +03:00
Wang, Yang
80d70dd438
Ywang2/update auto configuration api (#9977)
* Update AUTO OV 2.0 c++ configuration API.

Signed-off-by: ywang2 <yang4.wang@intel.com>

* Support the OV 2.0 key to set model priority and add the test case to verify the prioirty map logic within AUTO plugin.

Signed-off-by: ywang2 <yang4.wang@intel.com>

* Replace the old model priority key and add the corresponding test case.

Signed-off-by: ywang2 <yang4.wang@intel.com>
2022-01-31 16:24:46 +08:00
Nikolay Shchegolev
7d173eb354
[CPU] RNN, GRU, LSTM Sequences batch 1 tests failure. (#9719) 2022-01-31 11:21:39 +03:00
Mikhail Nosov
4e4b04bbd3
LSTMSequence/GPUSequence - fix validate_and_infer_types (#9996)
* Fix LSTMSequence/GPUSequence validation behavior consistent with RNNSequence

Fixed issue with no exception if num_directions=2, but 'm_direction' is not set to BIDIRECTIONAL. Previously there was no error with this (and luckily it failed later in some CPU transformations during compile_network)

Corrected several tests which use copy-pasted num_directions=2 without m_direction set
Also for dynamic 'num_directions' - output shape still has 1 or 2 directions, because m_direction is known. Tests for GRU/LSTM are updated for this
Also several tests worked incorrectly for LSTMv0 - expectation was specific error to be thrown, but no expection was also allowed

* Fixed clang-format
2022-01-31 08:24:43 +03:00
Tomasz Dołbniak
351c84e6e4
TestCase class fix (#10011)
* Store the expected output data in the TestCase class

* Skip the failing ONNX If tests

* Disable failing ONNX Softmax tests

* Disable the remaining failures
2022-01-31 08:11:18 +03:00
Chen Xu
49a01bf9fc
[CPU] Fix bug concerning model accuracy in Reduce node (#9929) 2022-01-30 17:36:36 +00:00
Chen Xu
cc5d30b26c
[CPU] Fixed TopK node optimization leftovers (#9862) 2022-01-30 19:11:47 +03:00
bell
aaf02ab4c7
modify optimal request num(add a multiplexer) (#10004)
* muliply the request# by 2

Signed-off-by: fishbell <bell.song@intel.com>

* modify case

Signed-off-by: fishbell <bell.song@intel.com>
2022-01-30 16:43:41 +03:00
Smirnov Grigorii
f424d2d8ae
Split transformations tests src to match IE transformations library structure (#55385) (#9949)
* refactor transformations directory

* create offline_transformations directory
2022-01-29 15:38:19 +03:00
Mikhail Nosov
2b87f000ed
ROI tensor support for Template plugin (#9914)
* ROI tensor support for Template plugin + tests for Template and CPU plugins

GPU doesn'tsupport ROI tensors, so tests were not added for GPU

* Added asserts for unsupported mixed axis order (like 0,3,1,2), and unsupported types like int4/int2 for ROI tensors
2022-01-29 10:52:52 +03:00
Edward Shogulin
1176b0f1f9
[LPT] MoveFakeQuantize parent output fix & refactoring (#9872)
* [LPT] MoveFakeQuantize parent output fix & refactoring

* [LPT] tests extending
2022-01-29 10:04:07 +03:00
Katarzyna Mitrus
e3ec1ac9b3
Allow duplicated outputs in ie_plugin_internal (#9882)
* Avoid duplicated outputs with the same name

* Revert onnx graph changes

* Allow output duplicates in ie_plugin_internal check

* Add test with onnx model

* Check get_tensor_ptr instead of any_name

* More outputs test

* Refactor to use std::transform

* test manifest update

* Remove redundant header

* INTERPRETER segfaults fix for duplicated output names

* Simplify duplication assert

* Update test names

* Test update
2022-01-29 02:58:09 +03:00
Egor Duplensky
9cb6626ffd
[CPU] Fix FullyConnected node for strided inputs and outputs (#9575)
FullyConnected node cannot work with strided inputs (for example
inplace descs) because it implicitly reshapes input and output tensors.
2022-01-28 22:20:09 +03:00
Maxim Andronov
b5ea943267
[CPU] MatMul dummy shapes creation fixed (#9858) 2022-01-28 21:21:02 +03:00
Anton Chetverikov
97a78d0059
[MO] EmptyTensorList transform (#9361)
* Initial change for new transformations

* Update patterns

* Update unsupported operation replacement

* Add input/output normalization passes call

* Update logic

* Refactor output concatination transform

* Update re_numerate_input_ports and shape infer functions for Loop

* Update comments

* Add back edge removing to output concatenation transformations

* Update comment

* Remove redundant normallization call

* Update supported layers list

* Use routine in check

* Add transformation to rub_before list
2022-01-28 20:53:16 +03:00
Ilya Lavrenov
eaa0a68fdb
Updated on OV 2.0 properties (#9906) 2022-01-28 18:52:50 +03:00
Sergey Lyubimtsev
4b35d48283
Add ability to generate requirements.txt files with dev tags (#9986) 2022-01-28 18:23:10 +03:00
Edward Shogulin
72583261fc
[LPT] Coverity issue fix: getMinValue (#9991) 2022-01-28 17:49:38 +03:00
Ilya Churaev
39b5cb809e
Fixed collisions in friendly names for internal operations (#9965)
* Fixed collisions in friendly names for internal operations

* Fixed renaming

* Added comments

* Renamed test file

* Fix behavior for outputs

* Fixed logic

* Fixed comments
2022-01-28 17:33:08 +03:00
Tatiana Savina
da1324db68
[DOCS] Enable sphinx panels directive (#9982)
* new directive

* add panels version
2022-01-28 17:18:57 +03:00
Alina Kladieva
8f3ebed8ec
Bump OMZ version (fix for pip-conflicts & wheels on Win & Mac) (#9987) 2022-01-28 16:45:28 +03:00
Gleb Kazantaev
45f98cece5
Load Time Improvements (#9946)
* Fix ConstantFolding; update ric fusion

* Update MarkPrecisionSensitiveDivdes pass

* Move back RIC fusion; Update copy_runtime_info logic + tests

* Update CF + tests

* Code style fix

* Code clean-up

* fix cf attr propagation

* Fix windows build
2022-01-28 16:36:45 +03:00
Maxim Gordeev
f0e2a2e811
Speech sample improvements (#9985) 2022-01-28 16:13:51 +03:00
Ilya Churaev
d5cd14efe3
Fixed cmake generation for CC build (#9974)
* Fixed cmake generation for CC build

* Fixed static build
2022-01-28 14:09:53 +03:00
Anastasia Popova
a61655040f
Fixed loader for TF2. (#9962) 2022-01-28 13:23:22 +03:00
Sergey Lyubimtsev
4cd20425c3
Building of openvino-dev wheel results many temp files in source tree (#9936) 2022-01-28 13:14:26 +03:00
Daria Mityagina
917994293c
[VPU] Added support for dynamic nodes for QueryNetwork - copy (#9925)
Currently, calling QueryNetwork from Myriad plugin with dynamic network could result in exception, this PR should fix this by removing nodes that could cause it from consideration.

Co-authored-by: Polina <polina.brzezinskaya@intel.com>
2022-01-28 13:06:37 +03:00
Alexandra Sidorova
2666a3dac4
[CPU] Enabled brgemm matmul primitives avx512 (#9824) 2022-01-28 11:34:32 +03:00
Szymon Irzabek
ef0a080323
[GNA] Extend matmul handling support (#9810)
Support matmuls with two non-const inputs.
Detect concat inputs to matmul as changing batch size and
handle appropriately.
Enable tests in GNA_SW_EXACT mode for convolution stride > kernel size.
2022-01-28 10:20:47 +03:00
Steve Yoo
068bdff787
Add SLT to Template Plugin: Loop-5 (#9789) 2022-01-28 07:56:04 +03:00
Steve Yoo
e8b954d6ec
Add testcases of index type i32 for NonMaxSuppression (#9902) 2022-01-28 07:53:01 +03:00
Wilson Seok
1e0470f4e7
Add slt in template plugin/lstm sequence (#9305)
* Remove fp16 of Convert layer test from skip_tests.config.cpp as it works now

* update repo

* initial code commit

* add runtime reference

* apply ov::Model

* initial lstmcell-1 definition

* initial change

* apply Peepholes

* apply input_forget option

* apply initial test case of lstmsequence-1

* fix clang-format error

* fix clang-format error 2

* add lstms_sequence test cases by runtime reference and onnx test cases

* fix clang-format error

* fix clang-format error

* fix onnx test failure of LSTM IE_CPU

* fix clang-format issue

* fix clang-format issue 2

* add type_prop and visitor api test of lstm_sequence_v1

* fix clang-format error

* replace input/refOut data to hard coded and remove unnecessary enum definition

* update namespace of Tensor()

* remove supported test cases in disabling list
2022-01-28 07:49:32 +03:00
Tomasz Dołbniak
658d9c3633
Removal of ONNX FE warning regarding Constant creation (#9935) 2022-01-28 07:16:18 +03:00
Yuan Hu
8883732ca8
[AutoPlugin] get optimal infer request number (#9529)
* dynamic optimal infer request

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* modify test case to match logic

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* for CPU_HELP,CPU case, use default value

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* update logic for GPU VPUX MYRIAD

throughput
GPU num = GPU_THROUGHPUT_STREAMS * 2
VPUX = 8
MYRIAD = 4

other mode
GPU NUM = 1
VPUX = 1
MYRIAD = 1

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* print GPU_THROUGHTPUT_STRREAM FOR TEST

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* get streams num from rangestreams for GPU

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* Revert "print GPU_THROUGHTPUT_STRREAM FOR TEST"

This reverts commit 06e348d21dfaed626bebcc32837737da9b036e44.

* use default variable instead of 4u and 1u

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* simple if else logic

Signed-off-by: Hu, Yuan2 <yuan2.hu@intel.com>

* check allow batching key

Signed-off-by: fishbell <bell.song@intel.com>

* add comment, optimize logic

Signed-off-by: fishbell <bell.song@intel.com>

Co-authored-by: fishbell <bell.song@intel.com>
2022-01-28 10:13:02 +08:00