IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
12,608 Commits 38 Branches 54 Tags 1.1 GiB
064a18eb09
Commit Graph

12608 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Pawel Raasz
93f2a1947c
Reduce binary size of CumSum implementation (#20004) 2023-09-22 20:41:32 +02:00
Pawel Raasz
efb8484124
[core] Api 2.0/migrate swish op to new API (#19934) 
* Migrate Swish operator to new API

* Remove `visit_attributes` as is same in base class
 2023-09-22 16:38:24 +04:00
Oleg Pipikin
ce378739f0
Refactor DetectionOutputLayerTest, DFTLayerTest, EltwiseLayerTest (#19922) 
* Refactor DetectionOutputLayerTest

* Refactor DFTLayerTest

* Refactor EltwiseLayerTest

* Fix

* Fix

* Disable tests
 2023-09-22 16:35:20 +04:00
Vladimir Paramuzov
5611f7d65c
[GPU] Fix constants reuse (#20003) 2023-09-22 15:29:38 +04:00
Mateusz Mikolajczyk
7a2ac27f09
[Spec][Opset13] BitwiseAnd-13, BitwiseNot-13, BitwiseOr-13, BitwiseXor-13 specifications (#19798) 
* [Spec] BitwiseNot-13 specification

* Improvements

* add other specifications

* Doc improvements

* Fix typos

* add opset13

* Fix typo

* add opset13

* Fix

* Improve example?

* Fix step formatting

* Apply suggestions from code review

Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>

* Add suggested changes

* Fix missing the

---------

Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
 2023-09-22 13:14:46 +02:00
Maciej Smyk
fb41324141
Update weight_compression.md (#20000) 2023-09-22 13:06:27 +02:00
Ekaterina Aidova
c76475288b
[PT FE]: support aten::scatter_reduce and extend aten::scatter (#19980) 
* [PT FE]: support aten::scatter_reduce and extend aten::scatter out arg support

* create mapping inside function
 2023-09-22 15:05:33 +04:00
Pavel Esir
9271b79540
[OVC] do not parse inputs for py_api (#19742) 
* [OVC] do not parse inputs

* fix unit-tests

* remove redundant lines, add test case

* add one more unit-test

* skip None values

* replace str with List in test_mo_import_from_memory

* corrected type hints, added a safety assert
 2023-09-22 15:05:21 +04:00
Oleg Pipikin
c1271d1217
Refactor EmbeddingBagOffsetsSumLayerTest, EmbeddingBagPackedSumLayerTest, EmbeddingSegmentsSumLayerTest (#19952) 
* Refactor EmbeddingBagOffsetsSumLayerTest

* Refactor EmbeddingBagPackedSumLayerTest

* Refactor EmbeddingSegmentsSumLayerTest
 2023-09-22 10:12:28 +00:00
Ekaterina Aidova
26d18c924b
[PT FE]: support aten::broadcast_tensors (#19994) 
* broadcast tensors

* [PT FE]: support aten::broadcast_tensors

* apply review comments

* remove add
 2023-09-22 13:54:44 +04:00
Maxim Vafin
2151e5f979
[PT FE] Add tests for timm models (#19997) 
* Add tests for timm models

* Apply suggestions from code review

* Some improvements

* Update tests/model_hub_tests/torch_tests/test_transformers.py
 2023-09-22 10:34:07 +02:00
Pawel Raasz
e7c1344d3c
[core] Api 2.0/migrate Add operator to new API (#19984) 
* Migrate Add operator to new API

* Remove `visit_attributes` as it calls base impl

* Use shape inference to calculate broadcast shape
 2023-09-22 11:57:02 +04:00
Pawel Raasz
aa293c09ac
[core]Api 2.0/migrate ceiling op to new API (#19909) 
* Migrate Ceiling op to new API

* Remove f16 precision from evaluate

* Correct ceiling reference doxy comments
- correct functor arg to be const
 2023-09-22 10:53:41 +04:00
Ekaterina Aidova
8d59fcd34f
[PT FE]: extend logical operations support (#19981) 
* [PT FE]: extend logical operations support

* tests

* more tests
 2023-09-22 10:11:36 +04:00
Egor Duplenskii
3f3d89678e
[CPU][ARM] Reorder FullyConnected weights in scope of compile_model (#19634) 2023-09-22 09:33:33 +04:00
Egor Duplenskii
6de8579b1d
[CPU][ARM] Choose eltwise layout based on model type (#19234) 
Eltwise order of the supported primitive descriptors affects the
performance of the model.
Often only one of the port descriptors matches with the layout
of the parent descriptors, i.e. when two parent ports have mixed
layout "nchw nhwc".
So either nchw or nhwc layout will be used for the eltwise node
and reorder will be used for either of the ports.
The shape of the ports also can be different (when one of the inputs is
broadcasted). So reorders on different ports have different
performance impact.
The layout of the eltwise node child has an effect on the performance
as well, since it may or may not require reorder on input.
 2023-09-22 09:33:16 +04:00
Egor Duplenskii
aec4c6c843
[CPU][ARM][FP16] Use Transpose executor for layout Reorder (#19227) 2023-09-22 09:32:12 +04:00
Ekaterina Aidova
fde054e4a6
[PT FE]: support aten::minimum aten::maximum (#19996) 2023-09-22 09:30:57 +04:00
Taylor Yeonbok Lee
f1b8abe55a
[GPU] Optimization for gemm & fc in iGPU. (#19780) 
* Optimization for gemm & fc in iGPU.
FC: fake alignment for 16 is better in iGPU.
Gemm: permute + gemm_tiled_opt is better than transposed_input + gemm_ref kernel for unaligned shapes to 16. Note that this is an temporal optimization and will be removed once the final solution (i.e., support unaligned transposed input shape in gemm_tiled_opt kernel) is availalbe.

* Fix unittest

* Fix for model_cache

* Fix unittest
 2023-09-21 22:07:53 -07:00
Pavel Esir
efe54362fd
fix f16/f32 el type mismatch for shape subgraphs: Parameter type should not be fused for precision sensitive nodes (#19959) 
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
 2023-09-22 02:06:39 +04:00
Georgy Krivoruchko
10c3b60aac
Updated model_creation_sample.py (#19989) 
Removed usage of ngraph in the sample
 2023-09-22 02:05:24 +04:00
Maxim Vafin
058b45e608
[PT FE] Fix aten::repeat regression (#19991) 
* Revert "[PT FE] Simplify repeat operation (#19926)"

This reverts commit f926e0e392.

* Fix aten::repeats regression

* Simplify

* Update src/frontends/pytorch/src/op_table.cpp

* Add impacted model
 2023-09-21 23:58:09 +02:00
Oleg Pipikin
b1bf16c7cf
Refactor ConversionLayerTest, ConcatLayerTest, ConstantLayerTest, ConvertColorI420LayerTest, ConvertColorNV12LayerTest (#19777) 
* Refactor ConversionLayerTest

* Refactor ConcatLayerTest

* Refactor ConstantLayerTest

* Refactor ConvertColorI420LayerTest

* Refactor ConvertColorNV12LayerTest
 2023-09-21 22:56:14 +02:00
Zlobin Vladimir
e068cfc5a3
benchmark: drop python3.7 (#19972) 
Ticket 118797
 2023-09-21 22:03:09 +04:00
Vladislav Golubev
0e0e1b0ee6
SmartReshape: ReshapeMatMul transformations are fixed (#19987) 
* SmartReshape: ReshapeMatMul transformations are fixed

* clang-format fixes
 2023-09-21 19:49:07 +02:00
Maxim Vafin
d0ef28e541
Reverse infer deduce second Squeeze input (#19923) 
* Reverse infer deduce second Squeeze input

* Fix build

* Fix tf tests

* Small optimization

* Fix type

* Fix tests
 2023-09-21 17:37:33 +02:00
Maxim Vafin
02f5b7a7e3
[PT FE] Add huggingface hub tests (#19858) 
* [PT FE] Add huggingface hub tests

* Update requirements.txt

* natten can't be installed

* Remove librosa

* Add xfail and skip marks

* Remove t5

* Free up space

* Apply suggestions from code review

* Report free space

* Clean pip cache

* Temporarily disable det2 models

* More du info

* Remove tmp info

* More du stats

* Fix du report

* Optimize du

* Ignore error

* Finalize changes

* Update tests/model_hub_tests/torch_tests/test_transformers.py
 2023-09-21 17:30:17 +02:00
Vitaliy Urusovskij
b00fbd04c5
Fix StressMemLeaksTests with several models (#19986) 
* Fix `StressMemLeaksTests` with several models

* Fix OMZ branch name in `get_testdata.py`
 2023-09-21 15:59:18 +02:00
Maxim Vafin
37d54bcb42
[PT FE] Support prim::TupleIndex operation (#19978) 
* [PT FE] Support prim::TupleIndex

* Update src/frontends/pytorch/src/op/tuple_index.cpp

* Update src/frontends/pytorch/src/op/tuple_index.cpp
 2023-09-21 15:57:42 +02:00
Maciej Smyk
f790a3b4f2
[DOCS] Python 3.7 support removal from docs for master (#19670) 
* Python update

* update

* fix

* m suffix removal
 2023-09-21 15:54:10 +02:00
Vitaliy Urusovskij
318106d17d
StridedSliceLayerTest and SqueezeUnsqueezeLayerTest to API2.0 (#19955) 
* `StridedSliceLayerTest` to API2.0

* `SqueezeUnsqueezeLayerTest` to API2.0

* Fix CppLint
 2023-09-21 15:08:51 +02:00
Tomasz Jankowski
c4adf80ec6
[Core] Move SlicePlan to Dev API (#19971) 
* Move SlicePlan to Dev API

* Recover legacy API

---------

Co-authored-by: Evgenya Nugmanova <evgeniia.nugmanova@intel.com>
 2023-09-21 14:41:01 +02:00
Vitaliy Urusovskij
e5acf880ad
Slice8, ShuffleChannels, ShapeOf layer tests to API2.0 (#19975) 
* `Slice8LayerTest` to API2.0

* `ShuffleChannelsLayerTest` to API2.0

* `ShapeOfLayerTest` to API2.0
 2023-09-21 14:36:26 +02:00
Vitaliy Urusovskij
fc9dedfd6a
Split, SpaceToDepth, SpaceToBatch layer tests to API2.0 (#19973) 
* `SplitLayerTest` to API2.0

* `SpaceToDepthLayerTest` to API2.0

* `SpaceToBatchLayerTest` to API2.0
 2023-09-21 12:29:30 +00:00
Ilya Churaev
e277d48535
Move unsqueeze evaluate to new API (#19948) 
* Move unsqueeze evaluate to new API

* Add new header

* Fixed comments

* Fixed build

* Revert redundant changes

* Fixed comments
 2023-09-21 16:07:10 +04:00
Vitaliy Urusovskij
2489356121
Skip failed TensorIteratorTest and CTCGreedy* CPU f16 tests on Mac (#19969) 
* Skip failed `TensorIteratorTest` CPU f16 tests

* Skip `CTCGreedyDecoder` on MacOS CPU ARM f16
 2023-09-21 15:44:37 +04:00
Siddhant Chauhan
d6d4e3deeb
Align openvino.compile_model and openvino.Core.compile_model functions (#19778) 
Co-Authored-By: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
 2023-09-21 11:15:30 +00:00
Nesterov Alexander
194b9f5c38
[ARM CPU] Migration Arm Compute Library to 23.08 version (#19523) 2023-09-21 14:30:38 +04:00
Pawel Raasz
5bf5e488b7
Migrate LogicalXor to new API (#19913) 2023-09-21 12:09:49 +02:00
Pavel Durandin
d90667c190
[GPU] Coverity fix (#19968) 2023-09-21 12:35:57 +04:00
Katarzyna Mitrus
7204b0b78d
[Spec][Opset13] NMSRotated-13 specification (#19574) 
* Init rotated non max suppresion spec

* Add opset13 docs

* Apply minor refactor from review

* Update boxes definition

* Update example format from cpp to xml

* Add version in op list

* Add clockwise attr to the example

* Align indent

* Remove redundant input from example

* Add steps of iou_rotated

* Add default values for attributes

* Drop box encoding attribute

* Rephrase nput description

* Applay grammatical suggestions
 2023-09-21 09:22:22 +02:00
Wilson Seok
c5843cf5d6
[GPU] use input memory buffer as output memory when input1/2 are empty (#19786) 
* use input memory buffer as output memeory when input1/2 are empty

* fix wrong rebase

* add func test

* implement in on_execute()

* remove deleted function definitioin

* remove unused header files

* fix include error

* update condition of empty input check
 2023-09-20 22:59:11 -07:00
Luwei Zhou
9dcd66c695
[CPU] Unify oc block to optimize peak memory. (#19575) 2023-09-21 09:54:55 +04:00
Roman Lyamin
a800d3e4f4
[GPU] Update weights reorder output shape for fully_connected (#19925) 2023-09-21 09:26:42 +04:00
Wanglei Shen
1786e366a2
Support new format for CPU information on Win7 (#19759) 2023-09-21 09:17:47 +04:00
Ilya Churaev
4a0e3fc77f
Remove ngraph from some sources (#19966) 2023-09-21 06:27:30 +04:00
Oleg Pipikin
69e9124eb7
Refactor ConvolutionBackpropDataLayerTest, ConvolutionLayerTest, DeformableConvolutionLayerTest (#19810) 
* Refactor ConvolutionBackpropDataLayerTest

* Refactor ConvolutionLayerTest

* Refactor DeformableConvolutionLayerTest

* Apply comments

* Apply comments

* Fix
 2023-09-21 00:50:52 +04:00
Evgenya Nugmanova
c1a8380052
Symbolic shape inference and graph optimizations (#19392) 
* Symbolic shape inference and graph optimizations
- Prepares a place in CommonOptimizations pipeline for symbolic optimizations
- Introduces symbolic propagation and symbolic optimizations for ChainedMaximum, NopBroadcast and shape sub-graph optimization
- Introduces utility runtime info for TableOfEquivalence passing and disabling of value invalidation during shape inference

* Executes NgramFusion in a symbolic environment. Relaxes Ngram fusion pattern utilizing symbolic knowledge

* Remove debug model visualization

* rt_info copying to new Add operation

* Fix visualization and place validation in nicer place in symbolic transformation

* Fix Slice operation not to propagate labels if input and output dimension is fully dynamic

* Covering Vladislav comments

* Replace value invalidation followed by validation to revalidation since it does the same thing

* Adding back invalidation of cached values to Symbolic Propagation pass

* Fix StridedSlice label propagation. Code style

* Update src/common/transformations/tests/symbolic_transformations/nop_broadcast.cpp
 2023-09-20 18:00:07 +04:00
Aleksandr Voron
8558476047
[CPU] [ARM] Enable SoftMax SLT tests on ARM (#19823) 2023-09-20 17:57:34 +04:00
Fang Xu
228ea44743
[CPU] Use omp section instead of task in UpdateNodes logic (#19831) 2023-09-20 17:47:48 +04:00