* [GPU] Shape agnositc optimized gemm kernel
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix CI failure
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Fix dynamic shape accuracy drop on SQuAD v1.1
- F1: 91.81%, EM: 85.25% @bert-small-uncased-whole-word-masking-squad-0001
Signed-off-by: Andrew Park <andrew.park@intel.com>
* Apply code review
Signed-off-by: Andrew Park <andrew.park@intel.com>
---------
Signed-off-by: Andrew Park <andrew.park@intel.com>
* [GPU] Fix the functional issue using fc:onednn in bert model.
* The issue had happened when input dims are 3 with post-po eltwise.
* oneDNN FC out supports 2-dims only, so OV need to update output and post-op too.
* Fix ACC issue in b16 onednn FC. cldnn updates yxfb format in b16 for opt kernel, but no need in onednn.
* Remove W.A code for running fc cldnn.
* Support gemm primtiive and multi types ForceImplTypes
* Change env name OV_GPU_ForceImplTypes
* Do not change elstwise post-op shape from original node: it caused the ACC issue when multiple users.
Signed-off-by: hyunback <hyunback.kim@intel.com>
* add support for aten::stack
* add new lines
* updated aten stack transformation
* add comments to the code
---------
Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
Co-authored-by: Andrei Kochin <andrei.kochin@intel.com>
* [TF FE] Refactor translators for Reduce ops and add layer tests
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
* Correct number of inputs
* Use already implemented logical operations tests
---------
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
+ Bugfix of eltwise_b_fs_yx_fsv16 kernel for int satuation
+ Add optimizing for fsv32, fsv16 using vload
+ Add optimizing for double blocked format eltwise
+ Support mixed format and broadcasting
+ Add test-cases to eltwise_gpu_test
Signed-off-by: Min, Byungil <byungil.min@intel.com>
* Added import model from memory for TF FE using string.
* Small correction.
* Clang format.
* Code correction.
* Implemented model importing to TF FE using temporary file.
* Removed wrong changes.
* Added check.
* Removed code duplication.
* Corrected logging of cli parameters.
* [CAPI][UNITE TEST] 1) create plugin test xml file in runtime; 2) fix arm plugin test error;
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [CAPI] fix format issue
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* [CAPI] change the time of Copyright
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
---------
Signed-off-by: xuejun <Xuejun.Zhai@intel.com>
* Remove "OV_FRONTEND_PATH" from `FrontEndManager`
* Remove "OV_FRONTEND_PATH" from layer_tests
* Remove "OV_FRONTEND_PATH" from find_ie_version.py
* Remove "OV_FRONTEND_PATH" from .ci/
* Move `ov::get_plugin_path` to file_util.hpp (ov::util)
* Add `register_front_end(name, path)` FEM public API
* Enable frontend/ tests
* ClangFormat
* Remove OV_FRONTEND_PATH from FE shared utils
* Add `register_front_end()` to PythonAPI
* Update Py tests with registering of "mock_py" FE
* ClangFormat
* Leftovers
* Fix SegFault with `register_front_end(name, lib_path)`
* Add FE_LIB_PRE and FE_LIB_SUFFIX to `ov_core_unit_tests`
* Fix format
* Update `testFailRegisterFEByWrongPath`
* Add eltwise types resolving. Support big int constants.
* Update src/bindings/python/src/openvino/frontend/pytorch/decoder.py
* Small fix
* Fix some cases
* Add tests for add in different types
* Add tests for mul
* Add tests for sub and div
* Small fixes
* Return list handling (needed for empty lists)
* Add test for empty list
* Update src/frontends/pytorch/src/op/mul.cpp
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
* Use refs instead of ptrs
* Apply suggestions from code review
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
* Apply code review suggestions
* Fix code style
* Add more eltwise ops
---------
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
fixes the FP32 and FP16 swap of results.
* SPR-6448Y updated results
Updated all files with data for SPR 6448Y. New results in benchmark-data.csv, benchmark-performance-data.xls, benchmark-info-detailed.xls and platform-list_22.3.pdf
* Use ov::Tensor in are_equal
* Use ov::Tensor in bounds evaluation
- remove from node evaluates bounds for HostTensor
- update validation utils
- update operators
* Fix code style
* Fix windows compilation errors
* Fix windows compilation errors in scatter update
* Improve reshape output shape calculation
during shape infer
* Add common bound evaluator
- rename evaluate_bound_t -> evaluate_bound
* Fix code formatting
* Move bound evaluation util functions to dev API
* Fix compilation issues
* Remove from dev API not required bound eval
- extract bound eval functions to separate source
* Add missing doxygen comment
* Use ref evaluate in bounds calc for shape_of
because input data can have dynamic type and
create tensor is not possible and data type has
no meaning for shape calculation.
