Commit Graph

799 Commits

Author SHA1 Message Date
Maxim Vafin
35e72251e9 [PT FE] Add support for aten::numpy_T and aten::feature_dropout (#20136)
* Add support for aten::numpy_t and aten::feature_dropout

* Update tests/layer_tests/pytorch_tests/test_transpose.py

Co-authored-by: Ekaterina Aidova <ekaterina.aidova@intel.com>

---------

Co-authored-by: Ekaterina Aidova <ekaterina.aidova@intel.com>
2023-10-03 09:52:29 +00:00
Roman Kazantsev
b409ea1930 [TF FE] Support TF1 While Control flow (#20105)
* [TF FE] Support TF1 While Control flow

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Apply code-style fix

* Update API for OpPlace to store back edge

* Fix build: no rvalue by reference passing

* Fix build issue: correct type

* Fix TF FE unit-tests

* Apply code-review feedback: remove unused vars

* Fix fusing complicated case of TF1 While

* Remove unused variable

* Update MO unit test

* Fix layer tests for While

* Handle Switch and NextIteration nodes connected directly

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-10-02 09:56:10 +04:00
Ilya Lavrenov
95e3096684 Added build on RedHat system to build & test RPM packages (#20134)
* Added GHA workflow for RPM packages

* Avoid rebuild for RPM / Debian packages

* Removed conditional include headers

* try only post-build

* Beautification

* Fixed testdata generation for mulit-config generators
2023-10-01 23:23:06 +04:00
Katarzyna Mitrus
86bf038417 [Ref][Core][Opset13] NMSRotated-13 core shell and reference implementation (#19907)
* nms_r_init

* Add tests

* Update nms refs

* Update onnx import

* Add nms rotated utils

* Remove soft sigma and align constructors

* Fix typo

* Style apply

* Add namespace for iou

* Update opset comment

* Revert cpu changes

* onnx test cleanup

* Update input types validation

* Align shape_infer boxes

* Test update

* Fix warning

* Temporary evaluate support for tests

* Add counterclockwise support

* Remove box_encoding attr

* Fix clockwise box idx

* More tests

* Update opset test

* Update boxes shape validation

* Type prop tests

* HostTensor to ov Tensor migration

* Update output_type set get output_type_attr

* Move setters and getters to cpp

* Add visitor test

* Cleanup

* Remove temp eval

* Headers adjustment

* use float for division

* Fix ref tests run

* Tests and style code refactor

* Move type check into box_last_dim

* Update visitor test namespace

* Check input type loop

* Remove nms_rotated namespace and rename ref function

* avoid copies in filling output tensor

* Update shape var name

* remove static from riou func

* Update nms_rot utils

* Move nms rot util to ov reference namespace

* use std::cos and std:::sin

* Update struct name

* Explain usage of postprocessing

* Update element type desc in error message

* Add more comments

* Adjust rotated util float types

* Fix name conflicts and warnings

* Update opset test ops number

* Move int input check to the loop

* Short box_def_size init

* Move remove static_output from shape_infer params

* Align float zero

* Update third-party-programs

* Fix TensorIt for CI

* Add op check test
2023-09-29 17:48:45 +00:00
Maxim Vafin
f38b5f4f06 [PT FE] Support moving TupleConstruct inside If body (#20081)
* Support moving TupleConstruct inside If body

* Fix win build

---------

Co-authored-by: Alina Kladieva <alina.kladieva@intel.com>
2023-09-28 23:11:09 +02:00
Maxim Vafin
84d98d8bf7 [PT FE] Add support for aten::pixel_shuffle (#20124)
* [PT FE] Add support for aten::pixel_shuffle

* Add comments

* Update src/frontends/pytorch/src/op/pixel_shuffle.cpp
2023-09-28 19:09:54 +02:00
Maxim Vafin
fea6db1a5f [PT FE] Fix issue when cat input is folded to tensor (#20090)
* [PT FE] Fix issue when cat input is folded to tensor

* CHeck real first input

* Update src/frontends/pytorch/src/op/cat.cpp
2023-09-28 10:52:09 +02:00
Ilya Lavrenov
132bada331 Enable PDPD tests build (#20078)
* Fixed comments for linux-riscv64 GHA workflow

* Try to enable PDPD tests back
2023-09-27 18:03:02 +04:00
Ilya Lavrenov
8349849e1c Use FindPython3.cmake (#19847)
* Use FindPython3.cmake

* Fixed compilation on macOS 14 with new core development tools

* Try to use Python3_SOABI instead of PYTHON_MODULE_EXTENSION

* Use Development.Module

* Keep specifying only Python3_EXECUTABLE

* Print PYTHON_MODULE_EXTENSION

* Added check for minimal cmake version for python API

* Returned Python3_INCLUDE_DIR for cross-compilation case

* Try to allow cmake older than 3.18

* Use build python interpreter to check cython dependency

* revert changes in .ci/openvino-onnx/Dockerfile

* removed unused code

* Fixed issue with variables scope

* Experiment: remove include dirs

* Corrected docs

* Use pybind11 function to set extension

* Revert "Experiment: remove include dirs"

This reverts commit 6f7f90211c.

* Refactor ConvolutionBackpropDataLayerTest, ConvolutionLayerTest, DeformableConvolutionLayerTest (#19810)

* Refactor ConvolutionBackpropDataLayerTest

* Refactor ConvolutionLayerTest

* Refactor DeformableConvolutionLayerTest

* Apply comments

* Apply comments

* Fix

* Updated minimum cmake version for Windows

* Simplified check

* Removed useless message status

* Use puiblic option

---------

Co-authored-by: Oleg Pipikin <oleg.pipikin@intel.com>
2023-09-26 20:57:29 +02:00
Mateusz Tabaka
5384fe43df Add EnableShapeOfConstantFolding transformation (#19880)
* Add EnableShapeOfConstantFolding transformation

Transpose sinking (that is used in TF frontend) disables ShapeOf constant folding
which prevents some optimizations further in the pipeline.
This patch introduces EnableShapeOfConstantFolding that removes DisableConstantFolding
from ShapeOf nodes.

Ticket: CVS-118890

* add description

* review comments

* headers
2023-09-26 15:53:23 +02:00
Ilya Lavrenov
85145cd60c Refactored Linix GHA pipeline (#20021)
* Refactored Linix GHA pipeline

* Move ONNX tests to Unit

* Install tests together with openvino

* Added checks for GNA usage in docs python snippets

* Install openssl for working pip

* Disabled onnx_dyn_shapes_expand_1_dyn_shape ONNX test

* exclude dyn shapes from onnx

* Migrated samples tests

* Fixed ONNX tests

* caches for PT tests dependencies

* Build Contrib separately from main OpenVINO

* More fixes

* Added python install for samples stage

* Installed manifests for ONNX FE tests

* Try to avoid libgl1 dependency in tests

* Clean-up

* Fixed issue with manifests in ONNX tests
2023-09-25 17:00:30 +04:00
Katarzyna Mitrus
37271c1e7c [Opset13][CPP] Initialization of Opset13 (#20005)
* Opset13 init

* Update test pattern ops

* Update remaining files
2023-09-25 10:20:37 +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
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
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
Ekaterina Aidova
fde054e4a6 [PT FE]: support aten::minimum aten::maximum (#19996) 2023-09-22 09:30:57 +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
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
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
Ekaterina Aidova
2c88fbf798 [PT FE]: support mixed precision in aten::min/max (#19936)
* [PT FE]: support mixed precision in aten::min/max

* fix eltwise dtype alignment for float16
2023-09-20 11:26:18 +04:00
Anastasiia Pnevskaia
215a2f435b tf.Graph decoder accuracy fixes (#19903)
* Fixed outputs passing from decoder.

* Fixed get_variable method.

* Code style.

* Removed xfails from precommit models.

* Minor fix.

* Comment fixed.

* Added test.
2023-09-19 19:04:24 +02:00
Maxim Vafin
f926e0e392 [PT FE] Simplify repeat operation (#19926) 2023-09-19 16:02:46 +04:00
Roman Kazantsev
d90ceb93d1 [TF Hub][TF FE] Fix 5D case for FusedBatchNorm (#19904)
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-09-18 12:51:33 +00:00
Evgenya Nugmanova
2205f48c71 Squeeze w/o axes [1, -1] -> dyn_rank (#19593)
Removes PDPD logic from core code, keeps PDPD specifics in the translator
2023-09-15 12:27:12 +04:00
Ilya Lavrenov
ba67db66ae Properly enable CMAKE_COMPILE_WARNING_AS_ERROR (#19828)
* Properly enable CMAKE_COMPILE_WARNING_AS_ERROR_DEFAULT

* Properly enable CMAKE_COMPILE_WARNING_AS_ERROR
2023-09-15 01:20:00 +04:00
Ilya Lavrenov
35a0706dff Replaced several cmake utilities with new ov_ prefix (#19819)
* Replaced several cmake utilities with new ov_ prefix

* Replaced several cmake utilities with new ov_ prefix
2023-09-14 16:22:50 +04:00
Maxim Vafin
1a950f9e8d [PT FE] Torchvision NMS can accept negative scores (#19826) 2023-09-14 11:07:24 +02:00
Maxim Vafin
4f92676c85 [PT FE] Support aten::one_hot (#19779)
* [PT FE] Support aten::one_hot

* Apply code style
2023-09-13 20:37:47 +02:00
Georgy Krivoruchko
5eff59a2d0 [ONNX] Switched ONNX to 1.14.1 (#18359)
* Switched ONNX to 1.14

* Updated IR_VERSION in the tests

* Assigned an extended tests to issues

* Switched ONNX to 1.14.1

* Slightly relaxed requirements for ONNX 1.14.1 and updated conan.lock
2023-09-13 15:40:08 +04:00
Ilya Lavrenov
6a1d680f90 Partially fixed github issue 18274 (#19758) 2023-09-12 07:34:45 +04:00
Ilya Churaev
7becaf8494 Remove legacy API from common test utils (#19647)
* Remove legacy API from common test utils

* Fixed code style

* Fixed build

* Try to fix Windows build

* Fixed GNA build
2023-09-12 07:09:12 +04:00
Vitaliy Urusovskij
9f4e918ee2 Gracefully fail if test models weren't generated (#19705)
* Gracefully fail if test models weren't generated

* Add assert instead of return `nullptr`
2023-09-11 15:18:45 +04:00
Roman Kazantsev
37f61551a3 [TF FE][TF Hub] Use ConcreteFunc input and output signatures (#19690)
* [TF Hub][TF FE] Preserve outputs of ConcreteFunction from signature and their names

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix naming and complete TODO

* Apply code-review: extra assert to check input_signature

* Fix inputs for fw

* Fix input data preparation and import convert_model

* Correct variable detection among all inputs

* Handle special input and output signature

* Fix adjust_saved_model_names

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-09-10 03:46:01 +00:00
Ilya Lavrenov
8eb165021c Fixed compilation with C++17 on Windows (#19682) 2023-09-09 01:53:20 +04:00
Maxim Vafin
e83697ded4 [PT FE] Implement override_all_inputs, override_all_outputs (#19642)
* Implement override_all_inputs, override_all_outputs

* Apply suggestions from code review

Co-authored-by: Piotr Krzemiński <piotrkrzeminski1234@gmail.com>

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

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

* Resolve problem with self input

* Update place.cpp

* Update place.cpp

* Fix build

---------

Co-authored-by: Piotr Krzemiński <piotrkrzeminski1234@gmail.com>
2023-09-08 22:31:49 +02:00
Maxim Vafin
bacb83f8a2 Support aten::tile op (#19645) 2023-09-06 17:28:22 +00:00
Ilya Churaev
7a4e765600 Removed legacy API from common frontend and shared tests (#19583)
* Removed legacy API from common frontend and shared tests

* Fixed build
2023-09-06 12:24:18 +04:00
Ilya Churaev
5fd327ae30 Remove legacy API from IR frontend (#19582)
* Remove legacy API from IR frontend

* Remove dependency from inference dev API
2023-09-05 15:41:57 +04:00
Xiuchuan Zhai
1b5f428752 eliminate broadcast node in masked_fill (#19595) 2023-09-05 10:36:30 +04:00
Ilya Lavrenov
c3b0c531f7 Added tflite to 'predefined_frontends' list (#19599) 2023-09-05 09:33:35 +04:00
Mateusz Mikolajczyk
c46f6bf115 [PT FE] Add aten::swapaxes (#19483)
* Add aten::swapaxes

* Add comment

* Improve swapaxes tests
2023-09-04 13:04:28 +02:00
Ekaterina Aidova
90ef7096b9 [PT FE]: support PReLU (#19515)
* [PT FE]: support PReLU

* Update tests/layer_tests/pytorch_tests/test_prelu.py

* Apply suggestions from code review

Co-authored-by: Piotr Krzemiński <piotrkrzeminski1234@gmail.com>

---------

Co-authored-by: Piotr Krzemiński <piotrkrzeminski1234@gmail.com>
2023-09-04 07:36:42 +00:00
Roman Kazantsev
2cf8f2bc1f [TF FE][GitHub issue] Support Selu operation and add test (#19528)
* [TF FE] Support Selu operation and add test

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix layer test

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-09-01 09:09:58 +00:00
Ivan Tikhonov
38cf4764cb Transformations: API 2.0 transition part 2 (#19475)
* Transformation component API 2.0:part 2

* Refactoring

* fix build
2023-08-30 14:28:20 +02:00
Bo Liu
6b57360c55 Fix warnings for Paddle Frontend (#19476)
* Fixed warnings for Paddle Frontend

* fix CI linux-gnu-9 build fail issue
2023-08-30 12:08:08 +04:00
Xiuchuan Zhai
36b9de1f25 enable sin/cos && fix top_k_v2 (#17525) 2023-08-30 08:55:57 +08:00
Roman Kazantsev
928c75623b [JAX][TF Hub][TF FE] Support XlaConvV2 operation and add JAX test (#19466)
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-08-29 12:28:12 +04:00
Maxim Vafin
c5b64e458b [PT FE] Align bool types and same bit int types (#19399)
* [PT FE] Align bool types and same bit int types

* Fix max value
2023-08-24 22:30:59 +02:00
Roman Kazantsev
8df85badf8 [TF Hub][TF FE] Support TensorListLength and TensorListResize operations (#19390)
* [TF Hub][TF FE] Support TensorListLength and TensorListResize operations

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add test with empty tensor list

* remove assert

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
2023-08-24 17:54:13 +02:00
Maxim Vafin
4882ccde03 [PT FE] Fix issue when FakeQuantize is not inserted after regular operations (#19314) 2023-08-22 17:18:23 +02:00