IntenseWebs/openvino
3
0 Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
13,725 Commits 38 Branches 54 Tags
032ac898e21c4008fb8c8dc2da5778c1b3ab4afe
Commit Graph
761 Commits
Author SHA1 Message Date
Surya Siddharth Pemmaraju bb0e4f8ecf Fixed issue with cat in fx backend (#20744) 
* Added fix for cat in torchfx

* Added batch_norm_legit_no_training op

* Fixed coding style

* Fixed clang format

* Addressed PR comments
 2023-11-01 08:36:19 +01:00
Przemyslaw Wysocki 31f6a95122 [PyOV] Set inference precision explicitly to fp32 for torchvision tests (#20713) 
* Set fp32

* Fix flake

* use properties api
 2023-10-30 20:06:55 +01:00
Anastasia Kuporosova 5b583276b2 [PyOV] Add get_byte_size to const node (#20759) 2023-10-30 19:46:07 +01:00
Ekaterina Aidova c21fd2ea61 [PT FE]: fix object has no attribute 'dtype' (#20755) 
* [PT FE]: fix  object has no attribute 'dtype'

* add model in tests

* update test
 2023-10-30 18:36:08 +00:00
Maxim Vafin 7b9db3d81b [PT FE] Add torch.int16 dtype support (#20735) 
* Add torch.int16 dtype support

* Add test
 2023-10-30 10:12:55 +01:00
Katarzyna Mitrus 14d51de93c [ONNX] Add support for standard ONNX GroupNormalization (#20694) 
* Add support for ONNX GroupNormalization

* Enable onnx compliance tests

* Add f to the test values to avoid cast warning

* Use tolerance as fp instead of bits
 2023-10-27 09:24:07 +04:00
Anastasia Kuporosova 3b9606f217 [PyOV] fix set_node_friendly_name (#20695) 2023-10-26 14:35:57 +02:00
Ilya Lavrenovanddependabot[bot] 0d458f086d Dependabot/pip/src/bindings/python/paddlepaddle 2.5.2 (#20708) 
* Bump paddlepaddle from 2.5.1 to 2.5.2 in /src/bindings/python

Bumps [paddlepaddle](https://github.com/paddlepaddle/paddle) from 2.5.1 to 2.5.2.
- [Release notes](https://github.com/paddlepaddle/paddle/releases)
- [Changelog](https://github.com/PaddlePaddle/Paddle/blob/develop/RELEASE.md)
- [Commits](https://github.com/paddlepaddle/paddle/compare/v2.5.1...v2.5.2)

---
updated-dependencies:
- dependency-name: paddlepaddle
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

* Removed WA for Ubuntu 22.04 and PDPD

---------

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2023-10-26 14:06:34 +04:00
Mikhail Ryzhov 49c5099f5e [GHA] Removed dependency on git repo in python tests (#20390) 
* Removed repo deps in py tests

* fixed typo

* changed test cmd

* add test_utils to pythonpath

* set test_utils to LD_LIBRARY_PATH

* changed path

* changed path

* source python site-packages

* fixed LD_LIBRARY_PATH

* print pip dir

* changed python install path

* source all openvino dir

* isolate test_utils

* system deps

* reordered deps

* dep conflict

* tests reorder

* changed test_utils path

* cpack test dir

* rebase fix

* fixed env

* frontend tests fix

* removed already installed reqs

* debug

* debug 2

* added python action path

* test

* reset 3rdparties

* set pip install path globally

* reset 3rdparties

* removed parallel code

* moved doc snippets

* switched back 4-core runner

* removed global unused vars

* removed unused requirements.txt

* removed unused requirements

* test

* Revert "test"

This reverts commit bdf22d22fd6ec5d488930bc362bdb0741c5bea76.
 2023-10-25 20:16:01 +00:00
Ilya Churaev 7ceff55b71 Add AlignedBuffer to OpenVINO developer API (#20532) 
* Add AlignedBuffer to OpenVINO developer API

* Fixed build

* Fixed code style and remove opset deprecation

* Fixed Windows build

* Fixed GNA

* Fixed comment
 2023-10-24 06:13:23 +00:00
Maxim Vafin 8d0381b0fe [PT FE] Implement custom op for types alignment (#20431) 
* [PT FE] Implement custom op for types alignment

* Fix code style

* Fix inplace ops

* Fix layer tests

* Remove no longer needed change

* Fix ovc tests

* Fix fe tests
 2023-10-23 22:54:08 +02:00
Andrey KashchikhinandIlya Lavrenov b67cff7cd5 [CI] [GHA] Introduce macOS ARM64 as a matrix parameter in the macOS pipeline (#20363) 
* add m1 mac pipelines as a matrix parameter

* Update mac.yml

disable java_api because of macos arm64 - Java is not available on macOS arm64 runners

* Update mac.yml

added always condition for all tests

* Update mac.yml

* Update mac.yml

* Update mac.yml

* Update setup.py

temp commit

* Update tools/openvino_dev/setup.py

* use matrix for var

* add mxnet to extras only for x86_64

* skip failing tests

* use xfail for Python tests; add missing filter for transformations tests

* skip CPU func tests on x86_64 mac; skip some tests from CPU func tests on arm mac

* Update mac.yml

* skip tests on mac arm

* skip tests on darwin; apply review

* add more skips for python and c++ tests

* skip tf tests

* skip more tf tests; skip more Python UT stages

* rm alwayses, rm triggers, add nightly trigger

---------

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-10-23 15:06:22 +04:00
Lucas Alava PeñaandJan Iwaszkiewicz fc93262d7c fix typo in opset3 shuffle_channels docstring (#20158) 
Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
 2023-10-20 15:00:50 +04:00
Mateusz MikolajczykandMaxim Vafin 891f79ac84 [PT FE] Add aten::as_strided (#19482) 
* Add aten::as_strided

* rm commented code

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

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>

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

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>

* Fix CI error

* Fix CI issues

* mark_node for remaining constants

* Add test reproducing issue

* Use strides from torchscript

* Add led model to test suite

* Add sugested changes

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-10-20 14:24:10 +04:00
Mustafa CavusandMaxim Vafin 3d5fe8d446 Llm and sd additional ops (#20435) 
* TorchFX: New ops added (baddbbmm, leaky_relu_)

* TorchFX: Initial scaled_dot_product_flash_attention

* Code Formatting: scaled_fot_product_attention translation

* TorchFX unit test enabled for SDPA

* Typo fix in comment line

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-10-19 21:21:28 +04:00
Mustafa CavusandJan Iwaszkiewicz 5018be82c3 TorchFX: Constant value pass without copy (#20380) 
* TorchFX: Constant value pass optimization

* Replace op.Constant with make_constant in fx_decoder

* Using shared memory for constant value passing

Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>

---------

Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
 2023-10-19 19:35:10 +04:00
Anastasiia PnevskaiaandAndrey Kashchikhin be177021b0 Switching telemetry to opt-out and disabling telemetry in CI jobs (#20391) 
* Added disabling telemetry script to linux azure job, added debug checks to convert_model.

* Telemetry disabling.

* Disabling telemetry.

* Config corrections.

* Config corrections.

* Update .github/workflows/mac.yml

Co-authored-by: Andrey Kashchikhin <andrey.kashchikhin@intel.com>

* Debug output.

* Win config correction.

* Win config correction.

* Debug output.

* Debug output.

* Added turning off telemetry to onnx azure tests config.

* Corrected config.

* Removed debug output.

* Switch telemetry to opt-out.

* Removed not needed blank lines.

* Text correction.

* Debug output.

* Remove changes from CI configs, add CI var to dockers.

* Config correction.

* Debug output.

* Config corrected.

* Readme corrected.

* Config changed.

* Config changed.

* Debug output.

* Required version changed.

* Remove debug output.

---------

Co-authored-by: Andrey Kashchikhin <andrey.kashchikhin@intel.com>
 2023-10-19 15:26:21 +04:00
Sergey LyalinAlexander KozlovEvgenya NugmanovaMaxim VafinIlya Churaev
 		46935e0a00 Partial GPTQ int4 models conversion support + Swap nibbles in u4/i4 element type (#20371) 
* Reference implementation for u4 constant compression from pytorch model based on bitwise ops pattern

* Fixed order of 4-bit halfs in byte

* Switched PyTorch FE to dev mode: in case if model cannot be fully converted, give partially converted model with PTFrameworkNode's with a printed warning (normally would raise an exception in case).

* Moved u4 compression to utils_quantize. Implemented not-interleaved version of u4 compression

* Removed debug output

* Added aten::matmul to the list of exceptions in may_produce_alias as a workaround for gptq models

* Added patching for gptq models applied automatically in convert_model

* WA for an inssue with u4 with earlier convert to fp16

* U4 blocked repacking for gptq patched model layout

* Deleted obsolete u4 re-packing based on aten::cat. Fixed the resulting u4 constant shape. Removed debug output.

* Revert "Switched PyTorch FE to dev mode: in case if model cannot be fully converted, give partially converted model with PTFrameworkNode's with a printed warning (normally would raise an exception in case)."

This reverts commit 0ef1455e70.

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

* Check mask and shift values in u4 pattern. deque -> OutputVector for u4_compression_stack

* Convert to a given floating type instead of half in gptq patching. Better structured code.

* Code style fix

* Removed deque include

* Code style fixes

* Trailing space removed

* Fixed patched_forward and ts_decoder after unvalidated commits.

* Swap nibbles in u4/i4

* Better exception handling around jit.trace and gptq.patch_model

* Update src/bindings/python/src/openvino/frontend/pytorch/gptq.py

Co-authored-by: Alexander Kozlov <alexander.kozlov@intel.com>

* Update src/bindings/python/src/openvino/frontend/pytorch/gptq.py

Co-authored-by: Alexander Kozlov <alexander.kozlov@intel.com>

* Code style

* Revers int4 byte order

* Fixed core tests

* Fixed unguarded dynamic_cast result

Co-authored-by: Evgenya Nugmanova <eva.my.link@gmail.com>

* Fixed transformation tests

* Update src/bindings/python/src/openvino/frontend/pytorch/gptq.py

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>

* Prevent patching of non-gptq models

* Removed extra calling of quantized weights decompression patterns

* Better detection of supported AutoGPTQ models + more diagnostics

* Accurate diagnostics in case when aten::stack has multiple axes

---------

Co-authored-by: Alexander Kozlov <alexander.kozlov@intel.com>
Co-authored-by: Ilya Churaev <ilyachur@gmail.com>
Co-authored-by: Evgenya Nugmanova <eva.my.link@gmail.com>
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-10-18 21:06:17 +04:00
Ilya Lavrenov 00618a429b Relocatable OpenVINO Dev package improvements (#20461) 2023-10-16 12:54:30 +04:00
dependabot[bot] 6f6017724f Bump pytest-timeout from 2.1.0 to 2.2.0 in /src/bindings/python (#20310) 
Bumps [pytest-timeout](https://github.com/pytest-dev/pytest-timeout) from 2.1.0 to 2.2.0.
- [Commits](https://github.com/pytest-dev/pytest-timeout/compare/2.1.0...2.2.0)

---
updated-dependencies:
- dependency-name: pytest-timeout
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2023-10-15 03:14:14 +04:00
Ilya Lavrenov 0b78a388be Removed custom python usage in Azure pipelines (#20433) 
* Removed custom python usage

* Removed custom python usage
 2023-10-14 01:18:27 +04:00
Katarzyna Mitrus 2478cf7df3 [Opset13][pyAPI] Python API Multinomial-13 (#20400) 
* Init Multinomial op python API

* Add python tests for Multinomial op

* Update num_samples input description
 2023-10-13 10:50:02 +00:00
Ilya Lavrenov 56d74a82cb Relocatable developer package (#20327) 
* Merge Linux CC + static build + clang compiler

* Improvements

* Removed ie prefixes from cmake scripts

* Fixes for NPU

* Initial relocatable OpenVINO Developer package

* Improvements

* Try to fix

* improvements

* Export a lot of headers

* Removed NVIDIA pipeline; make it to be a job

* Fixes

* Fixes 2

* Try ilya-lavrenov repo

* Clean-up

* More imrpovements

* Even more improvements

* Override export, install

* Override export, install

* Disable pythonwheel generation for relocatable OV dev package

* Fixed issues with versions

* Fixed android build

* Fixed android build

* Fixed NPU build

* Update src/bindings/python/CMakeLists.txt
 2023-10-12 22:59:04 +00:00
Ilya Lavrenov d41e7fcc30 Revert "[CI] [GHA] Skip test_div_uint8_cpu on macOS only; unskip test_onnx/test_backend.py in GHA workflows (#20367)" (#20402) 
This reverts commit a3d6d0bca9.
 2023-10-12 03:00:14 +04:00
Roman Kazantsev 0bb6450398 [TF FE] Support TF 2.14 and add OnesLike translator (#20385) 
* [TF FE] Support TF 2.14 and add OnesLike translator

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

* Update tests constraints

* Update open_model_zoo

* Adopt TF Lite test to 2.14 TF

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

* Support TF Lite layer tests for diffrent TF versions

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-10-11 15:24:32 +04:00
Andrey KashchikhinandIlya Lavrenov a3d6d0bca9 [CI] [GHA] Skip test_div_uint8_cpu on macOS only; unskip test_onnx/test_backend.py in GHA workflows (#20367) 
* only skip test if mac

* unskip

* unskip trigger

* skip for onnx fe as well

* do not skip

* return skips and unskip test_backend in Python API 1.0

* rm pr trigger

---------

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-10-11 14:05:12 +04:00
Jan IwaszkiewiczandMichal Lukaszewski 6a326455b9 [PyOV] Propagate errors on imports from runtime module (#20273) 
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
 2023-10-11 11:53:34 +04:00
Ilya Lavrenov 0a0d0f0aa5 Revert "Switch telemetry to opt-out approach. (#20290)" (#20370) 
This reverts commit b630bffa14.
 2023-10-10 22:08:20 +04:00
Andrey Kashchikhin 0060b26b74 [CI] [GHA] [Azure] Unskip test_div_uint8_cpu, unskip src/frontends/onnx/tests/test_python/test_zoo_models.py in Win and Linux pipelines (#20366) 
* only skip test if mac

* unskip

* skip for onnx fe as well

* remove skips

* rm unused imports

* revert import
 2023-10-10 21:06:18 +04:00
Mateusz MikolajczykandKatarzyna Mitrus 670fff062e [Opset13][pyAPI] Python API BitwiseNot-13 (#20265) 
* Add pyAPI BitwiseNot-13

* Update src/bindings/python/src/openvino/runtime/opset13/ops.py

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

* Remove whitespace from ops.py

---------

Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
 2023-10-10 14:27:45 +02:00
Andrey KashchikhinandIlya Lavrenov 1454e77bbf [CI] [GHA] Introduce GHA macOS Pipeline (#20212) 
* start transferring

* start with samples

* start with initial two stages

* change name

* skip pytorch tests; rm unused comments

* rm setupvars sourcing; make test steps similar to those in linux pipeline

* add missing options and setupvars sourcing

* add skips for mac

* install wheels directly

* add deployment target

* add skips for pytorch layer tests; experiment with samples

* do not exclude files for archives; set rpath

* apply comments; rm unnecessary stages

* Update mac.yml

fixed MO Python API tests

* Update .github/workflows/mac.yml

* Update openvino.cmake

add LC_RPATH to libopenvino.dylib

* Update src/cmake/openvino.cmake

* Update CMakeLists.txt

reverted changes in samples build

* Update openvino.cmake

removed rpath changes

* add setupvars

* disable pr trigger

---------

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-10-10 14:43:09 +04:00
Anastasiia Pnevskaia b630bffa14 Switch telemetry to opt-out approach. (#20290) 
* Switch telemetry to opt-out approach.

* Minor correction.

* Small correction.
 2023-10-10 13:14:45 +04:00
Mateusz Tabaka 4a3ce48f7a CompressQuantizeWeights optimizations (#20025) 
* Optimize CompressQuantizeWeights transformation

- remove CoordinateTransform usage from FakeQuantize reference implementation
- move ZeroPointOptimizer functionality inside CompressQuantizeWeights
- compute scale and zero point in the same loop

Ticket: CVS-119273

* review comments

* clang format

* fix comments
 2023-10-10 13:10:05 +04:00
Mateusz Mikolajczyk ef0c10df6c [Opset13][pyAPI] Python API binary BitwiseAnd-13, BitwiseOr-13, BitwiseXor-13 (#20261) 
* add xor

* Add and and or

* Add ch srequested in review
 2023-10-10 09:04:06 +02:00
Maxim Vafin d07b2ccc5f [PT FE] Return dynamic shape when static shape exist in graph (#20289) 2023-10-10 07:59:46 +02:00
Ilya Lavrenov e30f75bb4d Rpath story (#20297) 2023-10-10 06:27:26 +02:00
Ilya Lavrenov ead4b8a0ec Moved cmake functions, variables to API 2.0 naming style (#20281) 
* Merge Linux CC + static build + clang compiler

* Improvements

* Removed ie prefixes from cmake scripts

* Fixes for NPU
 2023-10-09 22:30:32 +04:00
Ilya Lavrenov 5764b5c108 Use RPATH instead of RUNPATH for wheel package (#20239) 2023-10-06 13:50:44 +04:00
Ilya Churaev 7c847cecba Remove ngraph namespace from operations without evaluate (#20252) 
* Remove ngraph namespace from operations without namespace

* Try to fix build

* Additional fixes

* More fixes

* More fixs

* Fix reverse op

* Fixed tests

* Throw an exception if somebody tries to reallocate tensor

* Revert "Throw an exception if somebody tries to reallocate tensor"

This reverts commit 8e06d6d576.

* Remove python test

* Revert "Remove python test"

This reverts commit 37b12148d3.

* Changed evaluate model behavior
 2023-10-06 12:54:11 +04:00
Przemyslaw WysockiAnastasia KuporosovaJan IwaszkiewiczIlya Lavrenov
 		9fe6be21ea [PyOV] Drop Python 3.7 in OpenVINO Runtime (#19534) 
* Drop 37

* Fix linter

* Minor change

* Minor changes

* update add_lib_path

* Minor changes

* Update pypi pages

* Update pypi pages

* Fix linter

* Minor changes

---------

Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-10-06 08:28:27 +00:00
Mikhail Ryzhov a79c07b3a0 [GA] Enable win workflow (#19646) 
* Enable win workflow and cpu parallel tests

* removed test code

* update cache naming

* extended logs collections

* Revert "extended logs collections"

This reverts commit 0dd6620832.

* revert lost code during the merge

* missed dependencies

* enabled push trigger

* changed the cache key name

* skipped failed test

* fixed github action condition and added comments

* Update src/core/tests/check.cpp

* cache generation fix

* Apply suggestions from code review

* fixed python test configuration

* Revert "cache generation fix"

This reverts commit 0feab650fe.

* debug parallel tests

* Revert "Revert "cache generation fix""

This reverts commit e385b04410.

* Revert "debug parallel tests"

This reverts commit e4459472a7.

* fixed steps conditions

* concurrency updated

* fixed test skip condition on win

* review changes

* collect debug logs

* overwrite test list

* debug commit

* Revert "debug commit"

This reverts commit 8720b87c8f.
 2023-10-05 16:49:39 +04:00
Katarzyna Mitrus bdb13aa28d [Opset13][pyAPI] Python API opset13 init and NMSRotated-13 (#20204) 
* Opset13 init

* Update py API to use Opset13

* Add nms_rotated op to py API

* Add tests

* Add trailng comma

* Adjust blank spaces

* Add nms_rotated to init file

* Update tests

* Update style

* Adjust Optional keyword

* Update op docs with shapes

* Add Tensor rypes hints

* Update tests to import opset13
 2023-10-04 17:05:29 +02:00
Ilya Lavrenov 749ed9dec7 Added python search path for Conda (#19929) 2023-10-03 22:47:52 +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
andreyanufr b73b2502b1 Introduce NF4 data type (#19900) 
* NF4 reference initial commit

* Compilable version.

* Executable NF4.

* Fixed nf4 unpacking.

* 1) Fixed warnings with nf4.
2) Removed unused functions.

* Added one test for nf4.

* Fixed code-style errors.

* Fixed code-style errors.

* Fixed NamingConventionCheck errors.

* Fixed test with nf4.

* Fixed windows compilation.

* Fixed casting warning.

* Fixed incorrect changes.

* Changed order of elements in nf4 pack/unpack.

* 1) Made Convert only on direction nf4->other type.
2) Applied reviewers suggestions.

* Fixed code style.

* Fised code style.

* 1) Added array header.
2) Added Bitsandbytes to third-party-programs.txt.

* 1) Removed unused code.
2) Fixed style typos.
3) Revert submodule version.

* Added test for nf4 compression.

* NF4 test refactoring.

* Added cpp tests for NF4.

* Removed model compilation from NF4 tests.

* Reverted submodule version.
 2023-09-28 20:56:57 +04:00
Anastasia Kuporosova 744f5d4130 [PyOV] Fix docs and add test for Model._get_raw_address (#20109) 2023-09-28 12:06:20 +00:00
Anastasia Kuporosova c90bcbf5eb [PyOV] fix if_op return types (#20014) 
* [PyOV] fix if_op return types

* fix if

* other types

* test for if

* add tests, try fix loop, ti

* fix ci
 2023-09-28 09:35:54 +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
Sonder 5079c50bb8 Update the regular syntax in cmakelist (#20066) 2023-09-27 12:26:17 +04:00
Anastasia Kuporosova 9dfed28aed [PyOV] Make openvino properties behave as python property object (#20007) 
* properties as property

* working prototype

* another attempt

* fix for test

* cosmetic changes

* generate properties

* test upste

* add comments

* update submodules

* apply comments
 2023-09-26 21:38:53 +02:00