### Details:
- `OPENVINO_ASSERT` takes boolean expression as first argument. In order
to explicitly throw AssertFailure, need to specify `false`, non-empty
string will be treated as `true`
---------
Co-authored-by: Sergey Shlyapnikov <sergey.shlyapnikov@intel.com>
### Details:
- Add new binding for dynamic which accept integer as argument.
- created tests in `test_partial_shape` & `test_partial_shape_equals`
### Tickets:
- Closes#23024
---------
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
### Details:
- The macOS x64 and arm workflows are broken.
- One of the issues is that `apt` retries are set unconditionally in
various reusable workflows when `apt` is only available on Linux.
- There was no `Smart_CI` job requirement for the `Samples` job in the
ARM mac workflow leading to it being broken.
- Several tests started failing, they are skipped in this PR with an
appropriate ticket.
### Tickets:
- *132596*
---------
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
### Details:
- Extended `Core` class to interact with Remote API via `compile_model`,
`create_context` and `get_default_context`.
- Added `RemoteContext` and `RemoteTensor` classes to Python API.
- Added `ClContext` and `ClImage2DTensor` classes placeholders.
- Added `VADisplayWrapper`, `VAContext` and `VASurfaceTensor` classes.
- Added utility function to raise `NotImplemented` errors.
**TODO list (with cc's):**
- [x] Decide which `VADisplayWrapper` constructors stay.
`VADisplayWrapper(VADisplay device)` (which translates to `void*`) is
quite necessary to pass memory pointers from/to external libraries.
`VADisplayWrapper(std::string& device)` is not necessary as it
introduces additional build dependency on libva itself. However, there
are no standard libva bindings that expose such functionalities.
- [ ] Proposal: change of `create_tensor` to `create_device_tensor ->
RemoteTensor` to make difference clear while API expose
`create_host_tensor -> Tensor`. (cc @ilya-lavrenov @slyalin)
- [x] Question: move CL and libva related classes to specific submodules
`openvino.intel_gpu` or keep it as-is in flat namespace? (cc
@ilya-lavrenov @slyalin @akuporos)
- [x] Decide on how memory management should work. My draft proposal:
expose in future PR as "read-only" via custom bindings to `cl_mem` in
case of accessing `ClImage2DTensor`.
- [x] Add test cases.
### Tickets:
- *129468 and 129469*
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
### Details:
- Missed adding two ops to the ops list when implementation was already
present
### Tickets:
- *ticket-id*
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
Review of any part is more than welcome!
### Details:
- Add Core for Inverse Operation
- Register Inverse Operation in Python API
- Add Reference implementation for Inverse - LU decomposition with
partial pivoting
- Add CPU implementation for Inverse - Parallel LU decomposition with
partial pivoting
- Add tests - Reference, Visitor, Type prop, Shape Inference, Python,
Conformance, CPU, Template
### Tickets:
- 131143
- 131144
- 131145
- 131147
- 131148
**Details:** Fix bad cast for tf.TensorShape to ov.PartialShape. The
error occurred in mobile-object-localizer-v1 model.
**Ticket:** CVS-132455
---------
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
### Details:
- Improved op support for 2024.0 release
- Includes ops required for enabling customer model lightglue
- Also added ops for lora fine tuning
- Added support for options in aot autograd backend
- Added a new option to specify whether to use aot autograd
functionality
- Added a new option to specify decompositions that need to be added
- Added a new option to specify ops that need to be disabled from
executing on OpenVINO
### Tickets:
- https://jira.devtools.intel.com/browse/CVS-116702
- https://jira.devtools.intel.com/browse/CVS-120276
---------
Co-authored-by: ynimmaga <yamini.nimmagadda@intel.com>
### Details:
- Extend Gather op with String support for data input
(agreed to do it within the same op version, existing reference
implementation is common for GatherBase and ready to handle strings, ref
changes are not needed, just evaluate supported types updated)
- Enable fallback to the reference implementation from CPU plugin
- Extend Template and CPU plugin operator tests with string validation
(most of the changes are tests related)
### Tickets:
- 130688
### Details:
- FakeConvert update to use ov::element::Type
- The changes are backward compatible, FakeConvert constructors with
std::string are not removed
### Tickets:
- 130680
* [PT FE] torch.export support
* Apply code style
* Fix build
* Support torch <2.2
* Support more operations
* Update tests/model_hub_tests/torch_tests/torch_utils.py
* Support for model operations
* Support is_causal as kwarg for SDPA
* Update src/frontends/pytorch/src/op/addcmul.cpp
* Update tests/model_hub_tests/torch_tests/test_timm.py
* Support only decoder passed to convert_model
* Fix tests
* Update src/bindings/python/src/openvino/frontend/pytorch/fx_decoder.py
* Update src/bindings/python/src/openvino/frontend/pytorch/fx_decoder.py
Co-authored-by: Ekaterina Aidova <ekaterina.aidova@intel.com>
* Apply suggestions from code review
* Apply suggestions from code review
* Improve testing with caching model
* Apply suggestions from code review
---------
Co-authored-by: Ekaterina Aidova <ekaterina.aidova@intel.com>
* Migrate `FactoryRegistry` to API 2.0
- remove add custom factory
- remove FactoryAttributeAdapter
* Fix opset name in ONNX FE
* Corrections in include files
* Fix build issues
* Use OpSet in visitor tests instead FactoryRegister
* Integrate `FactoryRegistry` into `OpSet`
* Remove ngraph/factory_adapter.hpp
* Fix auto batch test warning for windows build
* [TF FE] Support TensorFlow 2.15
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Use lower version of pylint
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Update tools/constraints.txt
* Update tools/constraints.txt
---------
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Add decorator to apply affixes
* Add new version of FQ to use affixes
* Add testcases
* Fix codestyle
* Make decorator only work on autogenerated Nodes
* remove print
* Get rid of distutils
* get rid of LooseVersion
* try to get rid of clean and copy_file
* fix mo tests
* fix warning in mo
* Update src/bindings/python/wheel/setup.py
Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
* add packaging to requirements
---------
Co-authored-by: Jan Iwaszkiewicz <jan.iwaszkiewicz@intel.com>
* fixed list index out of range error in node_decoder.py
* Improved and Modified as per review feedback
- Added a Model in the precommit_models for precommit validation.
- Removed skip and xfail marrks for the issue (129232 IndexError list
index out of range for Squeeze)
- Added explicilty return None in node_decoder.py
* Update tests/model_hub_tests/tf_hub_tests/nightly_models
* Update tests/model_hub_tests/tf_hub_tests/nightly_models
* Update tests/model_hub_tests/tf_hub_tests/nightly_models
* Apply suggestions from code review
* Update tests/model_hub_tests/tf_hub_tests/precommit_models
* Update tests/model_hub_tests/tf_hub_tests/precommit_models
---------
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
* Allow set_shape for ROI tensor
* Change set ROI shape up to initial ROI size
* Fix test value
* Update ROI tensor expected error message
in python tests
* Minor changes in ROI tensor ctor
Anastasia Kuporosova