IntenseWebs/openvino
3
0 Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
13,806 Commits 38 Branches 54 Tags
master
Commit Graph
62 Commits
Author SHA1 Message Date
Ilya Lavrenov c79ae17bbf Removed GNA plugin from 2024.0 (#21800) 2023-12-21 13:39:06 +01:00
Pavel Esir 137180bce5 [tests] resolves skipped HF tests: 3rd batch (#21678) 
* resolves skipped HF tests: 3rd batch

* remove comments

* minor corrections

* replace example input *argc -> **kwargc

* use config instead of dict in example input

* add kwargs
 2023-12-15 19:58:12 +04:00
Maxim Vafin be7c70c8c4 [PT FE] Support aten::scatter_add (#21633) 2023-12-13 17:54:33 +01:00
Pavel Esir 20ad8b43a1 Enable dinov2-base (#21626) 
with the newest transformers dinov-2-base started to pass
 2023-12-13 13:52:52 +00:00
Maxim Vafin 34b2b66b69 [PT FE] Report unsupported operations with model list (#21607) 
* [PT FE] Report unsupported operations with model list

* Continue on error

* Disable more models

* Check if file exists

* continue of error

* Fix on failure

* Fix file path

* Apply suggestions from code review
 2023-12-12 21:25:27 +00:00
Maxim VafinandRoboreptile 0f6b9abee8 Support aten::lstm, aten::gru and aten::rnn_{relu,tanh} operations (#21579) 
* [PT FE] Initial LSTM impl

* Support aten::lstm operation

* Support aten::gru and aten::rnn_{relu,tanh}

* Apply suggestions from code review

* Fix build

* Simplify implementation

* More simplification

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

* Simplification

* Use default LSTM helpers

---------

Co-authored-by: Roboreptile <piotr.krzeminski@intel.com>
 2023-12-11 19:06:51 +00:00
Pavel EsirandMaxim Vafin 948fc265b4 [tests] resolve skipped HF models: 2nd batch (#21526) 
* resolve skipped HF models: 2nd batch

* commented out models with no info and less than 10 downloads

* Revert "commented out models with no info and less than 10 downloads"

This reverts commit a55861ed69.

* replace with self.image

* resolve todo

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

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

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-12-11 15:12:12 +04:00
Evgeny Kotov 90087dbb48 Use time limits instead of number iterations limits in GHA performance tests (#21407) 
* use perf_counter_ns to measure performance; meausure within time not count

* fixes

* remove nightly

* set 60 secs by default

* fix output html

* code review fixes

* fix runtime duration

* remove unneeded inference

* cleanup
 2023-12-07 11:53:27 +04:00
Ekaterina Aidova d1f72e2d01 add gptq model test in pytorch hub tests (#21399) 
* conflict

* refactor test

* fix for case if can not instantiate config
 2023-12-01 10:56:05 +00:00
Maxim Vafin db6aeb7a65 [PT FE] Add torchbenchmark models to GHA tests (#21401) 
* [PT FE] Add torchbenchmark models to GHA tests

* Fix teardown

* Fix requirements

* Update tests/model_hub_tests/torch_tests/torchbench_models

* Update tests/model_hub_tests/torch_tests/torchbench_models

* Update tests/model_hub_tests/torch_tests/test_torchbench.py
 2023-12-01 09:44:12 +01:00
Roman Kazantsev c1cfb00f81 [TF Hub] Fix typo in test file name (#21408) 
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-30 23:49:01 +00:00
Maxim Vafin fa7a4f44b5 [PT FE] Update torch to 2.1 in GHA tests (#21391) 
* [PT FE] Update torch to 2.1 in GHA tests

* Adjust requirements for natten

* Add upper bound

* Use sys.executable
 2023-11-30 17:48:40 +00:00
Pavel EsirandMaxim Vafin d5d9fd11b3 [tests] resolve skipped HF tests: 1st batch (#21305) 
* initial

* some corrections for the first batch

* corrected classes for generate, corrected xfail

* leave only models with correct example input

* remove an obsolete line

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

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

* update musicgen

* cleanup test_hf_transformers.py

* typo fix

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

* move to up: corrected xfail

* revert back accidentally deleted elif

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-11-30 14:37:41 +00:00
Maxim Vafin 3cae6e7aea [PT FE] Fix timm model list according to newest version (#21376) 
* [PT FE] Fix timm model list according to newest version

* Remove xfail for hf model passed in latest nightly

* Fix code for loading vit_h model

* Restore torchvision
 2023-11-30 08:17:21 +04:00
Maxim Vafin 007b6fd82c [PT FE] Add support for GFPGAN model (#21371) 
* [PT FE] Add support for GFPGAN model

* Remove logs

* Fix codestyle

* Add support for aten::normal
 2023-11-29 23:05:46 +00:00
Maxim Vafin f3fe6b3f00 [PT FE] Add EDSR models to GHA tests (#21350) 
* [PT FE] Add EDSR models to GHA tests

* Apply suggestions from code review
 2023-11-28 21:05:33 +00:00
Roman Kazantsev 7c590bf180 [Hub Tests] Apply Python code formatting (#21286) 
* Apply Python code formatting

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

* Apply Python code formatting

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

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-28 16:54:02 +04:00
Maxim Vafin e52f922d35 [PT FE] Unify hub tests, add marks for timm models (#21283) 
* Unify hub tests, add marks for timm models

* Fix hf tests
 2023-11-27 12:22:13 +04:00
Luo Cheng 91660b1c05 [CPU] Fix ScaledDotProductAttension accuracy problem (#21217) 2023-11-26 14:47:04 +04:00
Evgeny KotovRoman KazantsevIlya Lavrenov
 		965313ba0b [TF Hub][Perf Tests]Implement performance tests in precommit to compare read_model and convert_model paths (#21023) 
* begin

* fixes

* measure time and compare

* fix bug + update precommint models

* cleanup

* add wget to install requirements; add to linux.yml

* fixes + improvements

* output results to html

* remove unneeded code

* fix

* code review fixes

* store downloaded models in cache

* use 1000 runs

* use model paths from tf-hub tests

* use tf hub api to download instead of wget; measure some time stats

* small fixes

* remove unneeded files

* use own list of models

* remove uneeded if

* remove unstable models

* remove unstable models

* fix requirements

* code review fixes

* Update .github/workflows/linux.yml

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* fix round_num function

* remove unstable network

* Update tests/model_hub_tests/models_hub_common/test_performance_model.py

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* code review fixes

* build fix

* code review fixes

* code review fixes

* code review fixes

* Update tests/model_hub_tests/models_hub_common/test_performance_model.py

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* Update tests/model_hub_tests/performance_tests/test_tf_hub_perfomance_model.py

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* Update tests/model_hub_tests/performance_tests/test_tf_hub_perfomance_model.py

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* Update tests/model_hub_tests/models_hub_common/test_performance_model.py

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* code review fixes: use autoflake

* remove unneeded tensorflow_text

* Update .github/workflows/linux.yml

* Update linux.yml

Added TensorFlow_Hub_Performance_Models_Tests to final status check

---------

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-11-24 18:13:44 +04:00
Roman Kazantsev bb773137f1 [TF Hub] Move all nightly scope to Kaggle links (#21169) 
7 models are skipped since they are not moved there. Now they are not available.

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-20 13:33:21 +04:00
Roman Kazantsev 2928b6bfed [TF Hub] Move pre-commit to Kaggle model links (#21113) 
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-16 14:24:45 +04:00
Maxim Vafin 369831caef [PT FE] Use system model cache when running torch hub tests locally (#21047) 
* Use system model cache when running torch hub tests locally

* Update tests/model_hub_tests/torch_tests/test_torchvision_models.py

* Update tests/model_hub_tests/torch_tests/test_torchvision_models.py
 2023-11-14 11:43:06 +01:00
Roman Kazantsev 75c72157dd [TF Hub] Add yamnet with complex tensors into precommit (#21035) 
After fixing ov.Tensor creation for empty numpy array, we are ready to support yamnet

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-13 17:42:13 +04:00
Maxim Vafin cdd342ea49 [PT FE] Add ALIKED to model tests (#20899) 
* Add ALIKED to model tests

* Update tests/model_hub_tests/torch_tests/test_aliked.py

* Update tests/model_hub_tests/torch_tests/test_aliked.py
 2023-11-07 09:34:26 +01:00
Maxim Vafin e976e7b90c [PT FE] Add tests for Speech-Transformer (#20847) 
* Add tests for Speech-Transformer

* Update tests/model_hub_tests/torch_tests/test_speech-transformer.py

* Update tests/model_hub_tests/torch_tests/test_speech-transformer.py
 2023-11-07 08:32:22 +01:00
Roman Kazantsev 494a9cf9a9 [TF FE] Refine tests for complex tensors support (#20905) 
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-07 10:29:45 +04:00
Roman Kazantsev 4a6e6e64c5 [TF FE] Fix conversion of FILM of keras.Model format (#20825) 
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-11-02 12:46:57 +04:00
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 bc9b52ef7e Increase timeout for hub tests (#20764) 2023-10-30 18:27:00 +04:00
Ekaterina Aidova 53c9a0f3d4 update pytorch layer tests for torch 2.1 compatibility (#20264) 
* update pytorch layer tests for torch 2.1 compatibility
 2023-10-30 08:30:01 +04:00
Maxim Vafin f029ebb8e2 [PT FE] Fix mask2former model marks in tests (#20717) 
* [PT FE] Fix mask2former model marks in tests

* Use better machine

* Add more models

* Update .github/workflows/linux.yml
 2023-10-27 21:04:30 +04:00
Maxim Vafin 46f46c6cc6 [PT FE] Fix xmod model test (#20682) 2023-10-25 08:00:34 +02:00
Maxim Vafin b17d0fe7f5 Update model list for hf hub tests (#20485) 
* Update model list and install cpu torch

* Move to hub tests

* Update tests/model_hub_tests/torch_tests/requirements.txt

* Make pytorch mainteiners owners of torch tests
 2023-10-20 20:12:00 +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
Ekaterina AidovaandMaxim Vafin 222fbb1aec [PT FE]: support aten::fill_diagonal_, aten::fill (#20395) 
* [PT FE]: support aten::fill_diagonal_, aten::fill

* remove xfail

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

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

* Update tests/model_hub_tests/torch_tests/test_hf_transformers.py

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
 2023-10-18 10:58:54 +02:00
Maxim Vafin f3b61d77e3 [GHA] Filter timm scope for model hub tests (#20525) 
* Filter timm scope

* Apply suggestions from code review
 2023-10-17 23:01:50 +04:00
Maxim Vafin 0fec05ecf2 Mark models according to last validation and fix some problems in models (#20324) 
* Mark models according to last validation and fix some problems in models

* Update tests/model_hub_tests/torch_tests/hf_transformers_models
 2023-10-10 11:23:23 +02:00
Mikhail Ryzhov 7e7b6482aa [GHA][HF] Switch python nightly tests to github runners (#20286) 
* switched runners to GHA

* excluded timm models from nightly scope

* added sudo steps

* switched precommit too
 2023-10-07 02:22:32 +04:00
Maxim VafinRoman KazantsevIlya Lavrenov
 		6664164ce4 Fix results validation (#20246) 
* Fix results validation

* Remove pt quantization

* Apply suggestions from code review

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* Refactor requirements

* Revert "Refactor requirements"

This reverts commit f78fe3d774.

---------

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-10-05 18:31:06 +04:00
Maxim Vafin 64cc3a96bb Restrict version of pytorch-quantization (#20137) 2023-09-28 23:11:41 +04:00
Mikhail RyzhovandIlya Lavrenov f85a276782 [GHA] Azure pip cache (#20029) 
* switched PyTorch_Models

* fixed root issue

* upgrade pip

* set bash environments

* install pip deps

* added git installation

* added g++ deps

* added python-dev package

* get particular python-dev package

* fixed package name

* removed python-dev

* cython dep

* debug

* fixed inlude path

* test cache

* set home dir

* changed disk space calculation

* crosspipeline pip cache

* added tag for source builded deps

* immutable deps

* use commit id instead of tag

* Update .github/workflows/linux.yml

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>

* fixed logs upload condition

* removed pip upgrade

* changed  detectotron version

* Update tests/model_hub_tests/torch_tests/requirements_secondary.txt

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>

* changed wheel installation way

* reset onednn

* added cleanup workload

* cleanup test

* fixed reqs install

* test cache depending on pip version

* syntax fix

* env overwriting test

* test env output

* fixed set env step

* move pip cahc initialization to step

* changed pip ver regexp

* set custom pip cahce for each step

* cleanup code and enabled test cleanup job

* test cleanup 2

* speedup removal

* check existed dir

* Enabled cache for samples job

* removed debug job

* changed way how to get pip ver

* improved pip version extration

* revert omz

* switched PyTorch_Models

* fixed root issue

* upgrade pip

* set bash environments

* install pip deps

* added git installation

* added g++ deps

* added python-dev package

* get particular python-dev package

* fixed package name

* removed python-dev

* cython dep

* debug

* fixed inlude path

* test cache

* set home dir

* changed disk space calculation

* crosspipeline pip cache

* added tag for source builded deps

* immutable deps

* use commit id instead of tag

* Update .github/workflows/linux.yml

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>

* fixed logs upload condition

* removed pip upgrade

* changed  detectotron version

* Update tests/model_hub_tests/torch_tests/requirements_secondary.txt

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>

* changed wheel installation way

* added cleanup workload

* reset onednn

* cleanup test

* fixed reqs install

* test cache depending on pip version

* syntax fix

* env overwriting test

* test env output

* fixed set env step

* move pip cahc initialization to step

* changed pip ver regexp

* set custom pip cahce for each step

* cleanup code and enabled test cleanup job

* test cleanup 2

* speedup removal

* check existed dir

* Enabled cache for samples job

* removed debug job

* changed way how to get pip ver

* improved pip version extration

* revert omz

* review fixes

* remove  cache files using atime

* mtime->atime

---------

Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
 2023-09-28 22:21:32 +04:00
Maxim Vafin f735999e7b Improve testing for hf models (#20017) 2023-09-22 20:10:33 +00:00
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
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 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
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 c10b45fe9e [PT FE] Fix issue with http error when using torch.hub (#19901) 
* [PT FE] Fix issue with http error when using torch.hub

* Mark failing models as xfail

* Remove incorrect model names
 2023-09-18 17:04:39 +02:00
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
Roman Kazantsev df19699e3a [TF Hub API][TF FE] Support TF Keras Model OOB without example_input (#19892) 
* [TF Hub] Cover TF Hub use cases with adoption to OpenVINO

This is necessarily to demonstrate support of models programmed with TF Hub API
through OV notebooks.

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

* Preserve original keras input and output tensor names

* Add tests with TF Hub API models

* No KerasLayer handling

* Handle specific signature

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
 2023-09-18 12:18:05 +00:00