c6258294fe
[CPU] Support group beam search ( #21954 )
2024-01-09 13:13:40 +04:00
316853053b
[CPU Plugin][Functional Test] upgrade infer request related to 2.0 ( #21981 )
...
* [CPU Plugin][Functional Test] upgrade infer request related to 2.0
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU Plugin][Func Test] remove folder behavior/infer_request and clean skip_tests_config.cpp
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* Fix test error
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
---------
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
2024-01-09 12:06:00 +04:00
ca1b118536
[CPU Plugin][Shared Test] remove unused definitions format code ( #21891 )
...
* [CPU Plugin][Shared Test] remove unused definitions format code
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* Fix conflict
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* Fix conflict
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* fix review comments
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
---------
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
2024-01-08 16:56:12 +04:00
1935bba20c
[CPU] Update Node constant type when adding or removing parent edges ( #21944 )
2024-01-08 16:24:17 +04:00
c0d564a465
[DOCS] Hyperlink 23.3 update for master ( #21908 )
...
* link update
* link update
* link update
* Update dev_api_references.rst
* Update dev_api_references.rst
* Update apidoc.py
* notebooks links
* notebook link update
* filename update
* link update
* link update api
2024-01-08 12:40:54 +01:00
8c09df0f3e
[CPU][API 2.0] Remove legacy header files and utils functions ( #21928 )
...
* Resolve ie::erase_if and ie::details::Caseless
1. replace ie::erase_if
2. move ie::Caseless to ov::intel_cpu::Caseless
3. replace ie_system_conf.h with system_conf.h
2024-01-08 09:32:12 +04:00
f0d8269ec3
[CPU] Enable bf16/fp16 inference precision support for platforms with avx2_vnni_2 ISA ( #20486 )
2024-01-08 09:21:47 +04:00
9c373d3335
[CPU] Fix exception message when querying for unsupported property ( #21987 )
2024-01-06 15:09:44 +01:00
90ee1004a4
[CPU] Ensure uniquness of a parent edge child port ( #21965 )
2024-01-05 11:55:13 +04:00
058f0206e2
[CPU] Fix debug build ( #21959 )
2024-01-04 19:12:12 +04:00
05f65add08
Port and clean `behavior/` shared tests #1 ( #21860 )
...
* Remove `set_preprocess.cpp`
* Remove `preprocessing.hpp`
* Remove `locale.hpp` - ported to `CanCompileModelWithCustomLocale`
* Port `version.cpp` and remove legacy
* Revert shared `version.hpp`
2024-01-04 16:28:16 +04:00
9c60a96e2c
[CPU] Fix ARM build ( #21945 )
2024-01-04 06:53:59 +00:00
f14e1c4d35
[CPU] Refactor graph transformations interface ( #21847 )
2024-01-03 15:37:59 +01:00
e99af17296
[CPU][ARM] Set `arm64-v8.2-a` as a default architecture ( #21545 )
2024-01-03 16:39:13 +04:00
03a5be788a
[CPU][ARM] Added acl_depthwise_convolution support for f16 convolution ( #21817 )
2024-01-03 13:31:59 +04:00
92cb11ee66
[Snippets] Added PassConfig and PositionedPasses for PassPipeline ( #21382 )
2024-01-03 12:36:15 +04:00
1bd05addb5
[CPU] Fix Convolution + Sum fusing shape mismatch error ( #21832 )
2024-01-03 09:22:18 +04:00
b1033784a0
[CPU] Fix gen_pattern::AttrAny::equal_to ( #21890 )
2024-01-02 10:36:12 +04:00
b821dcc785
[CPU] Support 4D matmul to FullyConnected conversion ( #21491 )
2023-12-28 16:38:11 +04:00
b51c5c0a99
[Commit slider] Apppaths cashing ( #21253 )
2023-12-27 15:43:52 +04:00
7691244404
[CPU Plugin][Func Test] upgrade single_layer_tests/psroi_pooling to 2.0 ( #21866 )
...
* [CPU Plugin][Func Test] upgrade single_layer_tests/psroi_pooling to 2.0
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* Skip the error test
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
---------
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
2023-12-27 15:09:56 +04:00
1f894bdb17
[CPU] Move FakeQuantize up through data movement operations ( #20858 )
2023-12-27 13:52:04 +04:00
e0eea90b55
[Snippets] Propagate work amounts from Subgraph To LinearIR ( #21338 )
2023-12-27 10:56:02 +04:00
95e8726132
[CPU] Refactored Emitters directory ( #21504 )
2023-12-27 10:09:32 +04:00
9df13267e8
[CPU][Tools] Update dump_check.py and readme.md ( #21229 )
...
- better display the command line at the end of the README.md
- fix the runtime error of register_pass
2023-12-26 17:58:33 +04:00
36d0b69b50
Add more tests for model cache new feature ( #21822 )
2023-12-26 11:57:37 +04:00
2eaed979a4
Remove G-API ( #21131 )
...
* Remove G-API
* Apply comments
* Revert "Apply comments" - tests shared classes only
This reverts commit 8d07082341
2023-12-25 14:50:54 +01:00
f104359f81
[CPU Plugin][Func Test] upgrade LoadNetworkActualWithThrow to 2.0 ( #21855 )
...
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
2023-12-25 13:17:57 +01:00
9647830f47
[CPU tests] fix enforce bf16 error ( #21634 )
...
* [CPU tests] fix enforce bf16 error
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU tests] fix review comments
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU tests] fix review comments
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU Func Test] fix error
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
---------
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
2023-12-25 12:45:04 +01:00
59d09d9b79
[CPU] Rework Node::isConstant() ( #21765 )
2023-12-25 14:58:59 +04:00
ca5a69eea0
[CPU] API 2.0 migration leftover ( #21620 )
2023-12-25 14:01:15 +04:00
7db0f6ec7c
[CPU][ARM] Disable transpose as reorder path for non-f32 precision ( #21845 )
2023-12-25 12:51:44 +04:00
f9a1a844f6
Refactor make_fake_quantize ( #21828 )
2023-12-25 12:24:22 +04:00
26576b0a30
Rework makeConstant builder ( #21319 )
...
* Rework make_constant node builder
* Fix
* Apply comments
* Apply comments
* Fix
---------
Co-authored-by: Pavel Durandin <pavel.durandin@intel.com>
2023-12-22 14:57:05 +01:00
6b1d2f8978
Additional clean-up for legacy components removal ( #21841 )
2023-12-22 15:47:40 +04:00
1fd3399cdf
Shared `execution_graph_tests` to API2.0 ( #21718 )
2023-12-21 13:27:24 +01:00
12a9d55c3e
Use new builders instead of old ones in tests ( #21742 )
...
* Use new make_convolution instead of old one
* Use new make_group_convolution instead of old one
* Use new make_convolution_backprop_data instead of old one
* Use new make_group_convolution_backprop_data instead of old one
* Use new make_binary_conv instead of old one
* Remove makeVariadicSplit builder
* Use new make_activation instead of old one
* Use new make_eltwise instead of old one
* Use new make_embedding_bag_packed_sum instead of old one
* Remove augru builder
* Fix clang-format
* Fix
2023-12-21 11:26:39 +04:00
ca5bf95e26
Xuejun/behavior tests core threading ( #21775 )
...
* [CPU Func Test] upgrade preprocessing related tests to 2.0
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU Func Test] upgrade version related tests to 2.0
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* [CPU Func Test] upgrade core threading related tests to 2.0
Signed-off-by: Zhai, Xuejun <xuejun.zhai@intel.com>
* Revert "[CPU Func Test] upgrade preprocessing related tests to 2.0"
This reverts commit abc27eabd2
2023-12-21 11:04:06 +04:00
56df9bc75e
Test utils to API2.0 + leftovers ( #21731 )
...
* Move shared `snippets` tests to API2.0
* Clean test utils from API1.0
* Move tests leftovers to API2.0
* Remove commented
* ClangFormat
* Fix comment
* Remove `single_layer/` includes from ported CPU tests
* Remove extra `InferenceEngine` usage
* CppLint
* Remove `single_layer/` usage from fusing_test_utils
* Remove `single_layer/` usage from GPU func tests
2023-12-21 10:02:58 +04:00
119dc95de6
Revert "[op conformance] Made fixes to allign with accuracy validation ( #21347 )" ( #21778 )
...
This reverts commit 8f13219728
2023-12-20 11:41:27 +04:00
dc79c41fb8
Updated xbyak_aarch64 in onednn ( #21762 )
2023-12-20 00:13:17 +04:00
7b1074b275
Generalize model cache reusing ( #21492 )
...
* Generalize model cache reusing
* Use .as<std::AnyMap>() to convert std::string to std::AnyMap
* Store driver version for different devices
* Use driver version as gpu compiled model runtime properties
2023-12-19 16:50:05 +01:00
6246fb0c23
[CPU] StringMemory for the output tensor in the InferRequest ( #21746 )
2023-12-19 15:20:15 +04:00
8f13219728
[op conformance] Made fixes to allign with accuracy validation ( #21347 )
2023-12-19 13:43:40 +04:00
a3a3f0b55c
[CPU] Fix chatglmv2 accuracy issue and crash with input 1024 token in SPR ( #21360 )
2023-12-19 13:13:35 +04:00
058d488083
[CPU] FullyConnected acceleration with 4bit weights decompression on SPR ( #21723 )
2023-12-18 23:16:45 +04:00
18a9c772b4
[CPU] Plugin optimizations for LLM beam search via states ( #21642 )
2023-12-18 23:16:32 +04:00
554e7d6730
[CPU] Dynamically shaped model with string tensor output fails ( #21713 )
2023-12-18 14:44:36 +04:00
610e0fab5c
[Opset13] ScaledDotProductAttention-13 input validation improvements ( #21281 )
...
* Improve test naming to match op
* Initial draft for scaleddot shape infer
* Fix formatting
* Improve shape & type infer
* Add shape tests
* Fix issues in shape infer
* Fix type infer
* Improve type_prop tests
* Ignore attention when causal
* Fix template type shape infer
* Fix issues with scalar inputs + test improvement
* Fix `get_node_input_partial_shapes`
* Improve shape/type validation and tests
* Allow for broadcastable inputs
* Improve tests
* Add CPU shape infer test
* Add broadcast shape infer test
* Use const
* Use const &
* Improve query input handling
* Fix test issues
* Change broadcast rules
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2023-12-18 10:30:54 +01:00
d613bb0e65
[CPU][ARM] Fix build CPU ARM plugin ( #21722 )
2023-12-18 13:20:33 +04:00
Luo Cheng