29bb5571e4
[PyOV] Remote API support for Python ( #22883 )
...
### Details:
- Port of https://github.com/openvinotoolkit/openvino/pull/22629
### Tickets:
- *N/A*
2024-02-16 16:13:44 +00:00
e28c8dd193
[CPU] Dynamic quantization for compressed FullyConnected ( #22738 )
...
### Details:
- *This PR adds activations dynamic quantization support for compressed
FullyConnected*
- OneDNN PR: https://github.com/openvinotoolkit/oneDNN/pull/229
### Tickets:
- *[CVS-128361](https://jira.devtools.intel.com/browse/CVS-128361 )*
2024-02-14 17:31:59 +00:00
5150118e0c
[Ref][Core][CPU][Opset14] Inverse Operation ( #22649 )
...
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
2024-02-14 11:58:38 +00:00
b48b799aa8
[PyOV] Add missing is_shared flag in memory sharing scenarios ( #22734 )
...
### Details:
- Add missing flag propagation in containers dispatchers for memory
sharing scenarios.
- Expected boost in cases of already aligned memory storages with
`__array__` interface, i.e. PyTorch tensors.
- Added testcases.
### Tickets:
- *132092*
2024-02-09 17:02:01 +00:00
dfe0699503
Rename IE to OV ( #22733 )
2024-02-09 13:36:41 +00:00
a2818110cc
[Op][Ref][CPU] Extend Gather op with String support ( #22465 )
...
### 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
2024-02-06 06:00:39 +00:00
3fb591a0e6
[PyOV] Update imports for properties ( #22594 )
...
### Details:
- *item1*
- *...*
### Tickets:
- *ticket-id*
2024-02-02 13:00:07 +00:00
5079abbdb9
[Op] FakeConvert update to use ov::element::Type ( #22574 )
...
### Details:
- FakeConvert update to use ov::element::Type
- The changes are backward compatible, FakeConvert constructors with
std::string are not removed
### Tickets:
- 130680
2024-02-02 09:55:11 +00:00
2990586789
[Opset14] Initialization of opset14 ( #22603 )
...
### Details:
- *Initialization of opset14 to reduce number of repeated changes in
each operator*
### Tickets:
- *N/A*
2024-02-02 08:43:54 +00:00
6d2d2d8e28
[Core]remove ie plugin config ( #22479 )
...
* [Core] remove ie_plugin_config
* Remove ie_internal_plugin_config and ie_plugin_config
* Remove ov::caching_properties and ov::exclusive_async_requests
* Fix failure cases and resolve comments
* Fix build error
* Remove myriad custom_op.cpp
* Fix build warning
* Fix capi auto plugin test failures
---------
Co-authored-by: Pavel Durandin <pavel.durandin@intel.com>
2024-02-01 08:56:31 +00:00
831abf98f1
[PyOV] Fix for affix decorator, allow Output instances ( #22504 )
2024-01-30 07:20:13 +00:00
27746ce3a5
[PyOV] Adding prefixes and suffixes to FQ operator nodes ( #22327 )
...
* 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
2024-01-26 06:00:57 +00:00
cb6eb83cc1
[core] ROI tensor can set new ROI shape ( #22257 )
...
* 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
2024-01-22 11:57:25 +01:00
1545ad08c4
[PyOV] Add optional string decoding for InferRequest and CompiledModel inference calls ( #22106 )
2024-01-18 13:07:33 +01:00
6bd9055440
[PyOV] Align add_extension methods across Python APIs ( #22166 )
2024-01-17 15:34:06 +01:00
5048299d97
Update year in src/bindings/python folder ( #22186 )
...
* Update year in src/bindings/python folder
* fix ci
2024-01-16 18:58:06 +01:00
b3c2c386dc
[FP8] Implementation of FP8 element types (ov::element::f8e4m3 and ov::element::f8e5m2) ( #21608 )
...
* FP8 element types init
* Remove redundant union
* Replace using ov
* Update fundamental types
* Update class name check
* Update tests
* Remove redundant sign
* Expose f8 types in Python
* Add to py to dtype map
* Style alignment
* Align python style
* Update test values
* Remove constexpr from_bits to fix warning
* Add trivially construcitble asserts and common constrexpr
* Align python tests opset
* Update f8e4m3 <-> float conversion
* f8e5m2 class update
* Add f8e5m2 unit test
* Add to string conversion tests
* Rename class f8e4m3 -> float8_e4m3
* Rename f8e5m2 -> float8_e5m2
* Remove size() and to_string from float8
- size() can be replaced by compile time sizeof
- to_string can be replaced by std::to_string()
* float8 E5M2 remove unused constexpr value
* Fix union initialization and ncc style rules
* Fix test issues
* Use NaN from std::numeric_limits instead macro
- minor refactor of float8_e4m3
* Update nf4 usage in element_type.cpp
* Sync openvino.style with master
* Update f8e5m2 test
---------
Co-authored-by: Raasz, Pawel <pawel.raasz@intel.com>
2024-01-16 15:26:43 +01:00
bd5b3434e7
Added missing cache_mode declarations for python and c API ( #22094 )
2024-01-12 11:11:26 +01:00
a4ce11d6d8
[PyOV] Add `get_then_body` to If operator ( #22080 )
2024-01-11 10:29:14 +01:00
ffcc2962d5
Remove more GNA related code ( #21826 )
...
* Remove more GNA related code
* Remove DEVICE_GNA from tests
* Fix according to review comments
---------
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
2023-12-27 16:09:22 +00:00
62dda1bd3e
[PyOV] delete test models from tests ( #21821 )
2023-12-21 17:45:40 +01:00
d10f49441d
[PyOV] Remove deprecated py api ( #21675 )
...
* [PyOV][Draft] Remove deprecated py api
* clean up
* remove symbol
* fix failure
* fix ci
* remove shared_memory from arguments
* update tests
2023-12-21 16:57:22 +01:00
80618b0498
Drop POT ( #21805 )
...
* Drop POT
* Removed POT transformations
2023-12-21 16:46:37 +04:00
dc64268564
Remove ov::hint::PerformanceMode::UNDEFINED ( #21592 )
...
* Remove ov::hint::PerformanceMode::UNDEFINED
* Update for reviewer comments and build issue
* Fix build error - may be used uninitialized
* Update
---------
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
2023-12-20 21:15:26 +04:00
712e374519
[PyOV] Allow replacing of invalid bytes in string-based data ( #21761 )
...
* [PyOV] Allow replacing of invalid bytes in string-based data
* Fix linter
2023-12-19 23:25:52 +04:00
c59498be9c
[PyOV] Allow single inputs in form of lists of simple types ( #21734 )
2023-12-18 22:14:31 +01:00
98e8caad79
[PyAPI][Opset13] ScaledDotProductAttention-13 PyAPI improvements + tests ( #21422 )
...
* Improve ScaledDotProductAttention PyAPI
* Add tmp tests
* Improve PyAPI & add tests
* Fix test issue
* Fix pytest type
* Update src/bindings/python/tests/test_graph/test_scaled_dot_product_attention.py
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
* Improve const creation
---------
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
2023-12-18 12:35:14 +01:00
57b1916f8d
[PyOV] Remove restrictions on opset13.constant to align with opset1-12 ( #21674 )
2023-12-18 10:02:43 +01:00
c4a49a3987
Add reset_state method to Python API ( #21660 )
...
* Add reset_state method to Python API
* fix missprint
* codestyle
* extend python tests
* delete debug code
* resolve review comments
2023-12-18 11:26:53 +04:00
fc88e14e83
Update evaluate methods of ReadValue/Assign operations ( #21372 )
...
* Update evaluate methods of ReadValue/Assign operations
* Update evaluate methods of ReadValue/Assign operations
* Prepare IVariable interface
* Support initializing subgraph
* Add exception types to the edge conflict searching algo
* Add test for init graph edge conflict
* Remove ShapeOf from consumer nodes in memory conflicts detection algo
* update template plugin
* Fix template tests; synchronize reset flag between VariableState and VariableValue objects
* fix legacy (Assign/ReadValue v3 + CNN Network) tests
* fix python tests
* fix python tests
* resolve review comments, apply cpu patch
---------
Co-authored-by: Maksim Kutakov <maksim.kutakov@intel.com>
2023-12-16 11:26:39 +01:00
eff9ba76ba
[PyOV] String type support ( #21532 )
...
* WIP working Tensor with automatic casting
* Update infer functions and minor fixes
* Impl for multi-dim arrays with fixes for strides
* Fix strides for S kind and refactor tests
* Added str_data and bytes_data to Tensor, cleaning up the solution
* Add test of warning while using data property
* Allow lists as inputs for single Tensors, refactor infer code to return OVDict with decoded strings, some refactoring, tests
* Replace string with another invalid input
* Added bytes_str and string_str properties, clean up common part of bindings, added test cases
* Improve string element type to be created from numpy/python counterparts, refactor of common code, small improvements
* Add tests for types
* Remove print
* Small fix for tensors from pointers
* Small fix for asserts
* Add tests for data dispatcher
* Fix comments
* fix tests
* Fix edge-case for scalar-like values and unlock tests for data dispatcher
2023-12-15 21:14:55 +01:00
c691dc483d
[PyOV][torchvision] Fix Resize for non-square images and expose missing interpolation modes ( #21673 )
2023-12-15 13:43:47 +01:00
948b7405c7
[Python API] Added exception handling for 'None' model argument in save_model function ( #21611 )
2023-12-15 11:36:57 +01:00
0a21205ad4
[PyOV] move rest of start_async tests to test_async ( #21643 )
2023-12-14 15:51:42 +01:00
1aa0c0ca0d
[PyOV] OVDict as input for infer/call methods ( #21275 )
...
* [PyOV] OVDict as input for infer/call methods
* for copied inputs
* improvement
* add test
* apply comments
* add test assert
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2023-12-13 18:13:18 +00:00
3e42ddbde5
[PyOV] support Type as arg in convert op ( #21603 )
...
* [PyOV] support Type as arg in convert op
* add test
* fix linter
2023-12-12 18:21:47 +01:00
ba34fa77e8
[PyOV] Expose result op ( #21482 )
...
* [PyOV] Expose result op
* update docs with op.Result
* codestyle
* add test and fix flake8 errors
* add result test
* fix transformation tests
* update return type
2023-12-08 13:57:11 +00:00
4c67ab7381
Merge changes from upstream/master referencing - test_infer_request.py renamed to test_sync_infer_request.py and new f… #21436 ( #21516 )
2023-12-08 09:48:13 +01:00
fd0809ead4
[AUTO] Enable round robin policy for cumulative throughput mode of AUTO plugin ( #20439 )
...
* Add and implement the logic of property SCHEDULE_POLICY for MULTI plugin.
* Updated.
* Enable test case for schedule policy test.
* enable test case for property ov::intel_auto::schedule_policy.
* Update.
* Updated.
* Updated.
* Update.
* Update the lock logic here by considering the runtime fallback case.
* Update.
* Update.
* Update.
* Update default value of schedule policy to DEVICE_PRIORITY
* Enable the function test case for schedule policy.
* Add description for inference requests schedule policy within AUTO plugin cumulative mode.
* Updated.
* Python bindings for enum SchedulePolicy and property ov::intel_auto::schedule_policy.
* Update.
* Update.
* Update.
* Updated.
---------
Co-authored-by: Chen Peter <peter.chen@intel.com>
Co-authored-by: Wanglei Shen <wanglei.shen@intel.com>
2023-12-07 07:04:56 +01:00
ca87784a29
[PyOV] Missed variable API in Model class ( #21434 )
...
* [PyOV] Missed variable API in Model class
* fixes in model docstr and codestyle
* add test
* fix ci
* fix ci
2023-12-04 15:44:23 +01:00
44f7bf7e3f
[CPU] Multinomial implementation ( #20406 )
...
* [CPU] Temp save commit
* [CPU] Add initial CPU implementation of Multinomial
* [CPU] Add parallel implementation with mock randomuniform
* [CPU] Fix accumulate incorrect iterator provided
* [CPU] Add tests for multinomial
* [CORE] Add lost tests
* [CPU] Add dynamic shape inference and descriptors init
* [CPU] Revamp tests to multiple files
* [CPU/SPEC] Apply suggested changes
* [CPU] Fix test compilation issues, clang fix
* Update multinomial.cpp
* [CPU] Fix Incorrect Primitive Descriptor for multiple combinations
* [CPU] Change params to inputs in testing function
* [CPU] Fix dynamic shape inference tensor access error
* [CPU] Save stable version
* [CPU] Add template execute for different input dtypes
* [CPU] Introduce new method of loading data to tests, fix dynamic shape inference
* [CPU] Improve parralelism
* [CPU] Improve pararrelism - fix indexes
* [CPU] Fix no_replacement tests, fix randomness in tests
* [CPU] Split tests into log and no_log version to avoid rounding when values are close to 0
* [CPU] Add mersenne-twister seed and random_uniform distribution as source for randomness, add debug prints
* [CPU] Apply suggestions from review, fix 4x4 log tests
* [CPU] Force i32 convert format
* [CPU] Fix double to float conversion warning
* [CPU] Remove debugging prints, fix CIs float error
* [CPU] Fix for convert_type in CIs
* Update src/plugins/intel_cpu/src/shape_inference/custom/multinomial.hpp
Co-authored-by: Mateusz Mikolajczyk <mateusz.mikolajczyk@intel.com>
* Update src/plugins/intel_cpu/src/nodes/multinomial.hpp
Co-authored-by: Mateusz Mikolajczyk <mateusz.mikolajczyk@intel.com>
* Update src/plugins/intel_cpu/src/shape_inference/custom/multinomial.hpp
Co-authored-by: Mateusz Mikolajczyk <mateusz.mikolajczyk@intel.com>
* [CPU] Migrate to CPU API 2.0
* [Ref/CPU] Remove support for 1D tensors, use Core Shape Inference
* [CPU] Remove unnecessary symbols
* Update multinomial.cpp
* Update multinomial.cpp
* Update ops.py
* [CPU] Fix const identifier missing after reinterpret cast
* [CPU] Fix Mac cpplint error
* [CPU] Apply recommended changes - 0-seed nondeterminism, casts in testsshape_infer optimization
* [CPU] Apply iterator optimization suggestion
* [CPU] Replace casts with class constructors in tests
* [CPU] Remove unnecessary static_casts to void*
* Update multinomial.cpp
* [CPU] Apply suggestions from review - move template, fix i64 precision, redturn off shape precision for const inputs, set always-execute for const inputs
* [CPU] Relocate tests to shared, remove using namespace from header files
* [CPU] Add definitions for files eaten by clang fix
* [CPU] Fix seed for Mersenne Twister Engine
* [CPU] Try fix incorrect 1x3 for 3 samples test (bf16)
* [CPU] Use only mersenne for seed generation
* [CPU] Relocate test, add debug prints
* [CPU] Add relocated test that got eaten
* [CPU] Remove uniform distribution, replace with division by max value
* Update multinomial.cpp
* Update multinomial.cpp
* [CPU] Add explicit float cast for CIs
* Update multinomial.cpp
* [CPU] Use intel_cpu::bfloat16 to reduce innacuracies
* [CPU] Remove debug caps, all tests pass
* [CPU] Clang fix
* [GPU] Remove GPU 1D test case
* [CPU] Modify tests to add seed=0 case, add ignore statement for this test and add subtask to complete after current release
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
Co-authored-by: Mateusz Mikolajczyk <mateusz.mikolajczyk@intel.com>
2023-12-02 12:46:17 +01:00
0e642e984b
[PyOV] tests refactoring ( #21410 )
...
* [PyOV] tests refactoring
* remove ngraoh mentioning
* codestyle
* replace func
2023-12-01 14:53:33 +01:00
74bf3d4e38
Extend Python API with ScatterNDUpdate-3 ( #21325 )
...
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
2023-11-30 15:32:45 +00:00
bc78fd359e
[PyOV] Dynamic attributes for Python based operators ( #21256 )
2023-11-27 16:16:02 +01:00
57d794c810
ConvertPrecision transformation: handle Assign, ReadValue and Variable ( #21266 )
...
* ConvertPresicion transformation: handle Assign, ReadValue and Variable
* revert debug code
* use correct commit for onednn_gpu
* codestyle
2023-11-27 18:34:31 +04:00
ccfe58cf84
[PyOV] Update imports in py api tests ( #21252 )
...
* [PyOV] Update imports in py api tests
* update year and back test_compression_4bit.py
* missed import
2023-11-24 11:16:57 +04:00
a329611349
Support dynamic shapes in ReadValue/Assign ops and MakeStateful transformation ( #20404 )
...
* Support for dynamic shapes in ReadValue/Assign ops and MakeStateful transformation
* fix unit tests
* codestyle
* fix warning
* fix visitors test
* Fix lowLatency transformation, fix functional tests
* fix LowLatency2
* codestyle
* Fix CPU/GNA tests
* fix shape inference of assign operation
* fix python API
* fix mypy error
* Fix read_value operation version in python API for opsets7-13
* Update python API and fix tests
* set the same Varibable obj to Assign/ReadValue in py API, disable fp16 memory tests
* Update shape inference methods of Assign, ReadValue; update tests; fix codestyle
* fix python codestyle
* fix tests
* Resolve review comments
* codestyle
* resolve review comments
* codestyle
* Update ReadValue op in kaldi frontend, fix GNA tests
* fix Assign/ReadValue bindings, fix read_value test
* Apply suggestions from code review
Co-authored-by: cecilia peng <cecilia.peng@intel.com>
* Fix functional tests, add ov::Type support as an argument for read_value py binding
* temporary disable test_query_state_write_buffer fp16 tests
* Fix backward compatibility with the IRs with ReadValue/Assign ops generated via prev versions of OV; fix func tests
* fix python test, support type as str in read_value/assign bindings
* fix model binding
* fix pre-commit tests: variables vector cleanup
* fix integration with cpu changes
* codestyle
* fix warning
* Fix type/shape validation for read_value op
* fix a type_prop test for assign op
---------
Co-authored-by: cecilia peng <cecilia.peng@intel.com>
2023-11-21 20:59:36 +04:00
54a68c8a13
[Opset13][pyAPI] Python API FakeConvert-13 ( #21155 )
...
* [Opset13][pyAPI] Python API FakeConvert-13
* Fix typo
* Add experimental warning
* Add missing py opset in docs
* Fix flake
* Apply suggestions from code review
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
* Apply suggestions from code review
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
* Apply requested changes
---------
Co-authored-by: Anastasia Kuporosova <anastasia.kuporosova@intel.com>
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>
2023-11-21 14:25:02 +00:00
20af42cfd9
Add python bindings for PILLOW interpolation modes ( #21188 )
2023-11-20 22:39:05 +00:00
56301d8878
[PyOV] add param-result to offline transformation ( #21140 )
2023-11-20 15:59:02 +01:00
Jan Iwaszkiewicz