Cecilia/opt/host #136
Closed
Cecilia/opt/host #136
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
…4916) Current Gemm in dpas always compile and use onednn kernel in 1st inference. Shape agnostic Gemm kernel is not used. So PR enable SA Gemm in dpas for reducing 1st inference latency. ### Tickets: - *143315* --------- Signed-off-by: hyunback <[email protected]>
Adding reference to Hugging Face in documentation.
MVN SLT f16 tests were disabled by setting high threshold because of ACL accuracy issue: ARM-software/ComputeLibrary#1095 Issue is fixed and ACL with the fix is being used by OpenVINO
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Bumps [codecov/codecov-action](https://github.com/codecov/codecov-action) from 4.4.1 to 4.5.0. <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/codecov/codecov-action/commit/e28ff129e5465c2c0dcc6f003fc735cb6ae0c673"><code>e28ff12</code></a> chore(release): bump to 4.5.0 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1477">#1477</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/7594baaf0c8b9ecdddba6269a6b089f60a3c859a"><code>7594baa</code></a> Use an existing token even if the PR is from a fork (<a href="https://redirect.github.com/codecov/codecov-action/issues/1471">#1471</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/81c0a51dcc5a7dd6e63f9b1214da9715a842300d"><code>81c0a51</code></a> feat: add support for tokenless v3 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1410">#1410</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/f5e203f3d95eae0da58ec699083e7bcd7b8b69ae"><code>f5e203f</code></a> build(deps-dev): bump <code>@typescript-eslint/eslint-plugin</code> from 7.12.0 to 7.13.0 ...</li> <li><a href="https://github.com/codecov/codecov-action/commit/7c48363ed341bdfcfee0cc2d33b0743003134af2"><code>7c48363</code></a> build(deps-dev): bump braces from 3.0.2 to 3.0.3 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1475">#1475</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/69e5d09ca1b986185c1dd546857502976e50f5bc"><code>69e5d09</code></a> build(deps-dev): bump <code>@typescript-eslint/parser</code> from 7.12.0 to 7.13.0 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1474">#1474</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/feaf700b2a674720c51aa264a0558178b0ae4fed"><code>feaf700</code></a> fix: handle trailing commas (<a href="https://redirect.github.com/codecov/codecov-action/issues/1470">#1470</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/7b6a727a4ae8e34ed4b8c8cc2c0710bb57de27d2"><code>7b6a727</code></a> build(deps): bump github/codeql-action from 3.25.7 to 3.25.8 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1472">#1472</a>)</li> <li><a href="https://github.com/codecov/codecov-action/commit/ccf7a1fd43a718f28675f2aa46aed231bc5e6c5f"><code>ccf7a1f</code></a> build(deps-dev): bump <code>@typescript-eslint/eslint-plugin</code> from 7.11.0 to 7.12.0 ...</li> <li><a href="https://github.com/codecov/codecov-action/commit/f03f015db9dd13f118dbaff6724f3b9c2cdd6f7b"><code>f03f015</code></a> build(deps-dev): bump <code>@typescript-eslint/parser</code> from 7.11.0 to 7.12.0 (<a href="https://redirect.github.com/codecov/codecov-action/issues/1467">#1467</a>)</li> <li>Additional commits viewable in <a href="https://github.com/codecov/codecov-action/compare/125fc84a9a348dbcf27191600683ec096ec9021c...e28ff129e5465c2c0dcc6f003fc735cb6ae0c673">compare view</a></li> </ul> </details> <br /> [](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) </details> Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
…#20230) ### Details: - *Align OV with latest renaming in `vpux-plugin`* - After [PR openvinotoolkit#5738](intel-innersource/applications.ai.vpu-accelerators.vpux-plugin#5738) in `vpux-plugin` was merged, we are ready to update `ov::test::utils::DEVICE_KEEMBAY` to `ov::test::utils::DEVICE_NPU` on openvino side. ### Tickets: - *[EISW-89683](https://jira.devtools.intel.com/browse/EISW-89683)*
…25023) ### Details: - zero-backend/pci information: Older drivers return errors not covered by current error managment, leading to Throw in zero_device intitialization, leading to total incompatibility with certain old drivers. Since pci information is not essential for the correct functioning of the plugin, i am removing throws completely, falling back to error message and default values. ### Tickets: - none
…24327) To reduce infer overhead
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
…olkit#25011) ### Details: - *[LPT] ReduceSum: zero point support* - *Original PR: openvinotoolkit#24977 ### Tickets: - *CVS-142256*
### Details:
- For dims A, B, C, D, abcd format is expressed in strides: {B*C*D, C*D,
D, 1}.
abdc format would be expressed in strides the following way: {B*C*D,
C*D, 1, D}.
Adjust strides with B_, C_, D_ accordingly to make them bigger for
certain dimensions.
for example, input 1 dimension as [2,3,11,7], transposed input strides
should be 231x77x1x7
./benchdnn --matmul --engine=gpu.0 --stag=abcd ---strides=:231x77x1x7:
2x3x5x7:2x3x7x11
### Tickets:
- 142250
---------
Signed-off-by: song bell <[email protected]>
Co-authored-by: Chen Peter <[email protected]>
…olkit#25043) ### Details: - skip random error in KV Cache test (This is to fix original test, no new test needed) ### Tickets: - 142899
…support (openvinotoolkit#24999) ### Details: - Add axis attribute to crop for `VariadicSplit` and `Split` dynamic support - This attribute is required to apply dynamic padding by `prepare_buffer_fusing` opt pass at build time ### Tickets: - 143096 --------- Signed-off-by: Andrew Park <[email protected]>
### Details: - *crash when compile_model with model path and ov::cache_dir is set* - *...* ### Tickets: - *N/A* --------- Co-authored-by: Chen Peter <[email protected]>
### Details: - Run tests 1000 times and can't reproduce it ### Tickets: - CVS-123019
…fix dependency issue (openvinotoolkit#24926) ### Tickets: - *134419* - *134423*
### Details: - Add human-pose-estimation sample for js bindings ### Tickets: - 141204
…thon (openvinotoolkit#25056) Updates the requirements on [flake8](https://github.com/pycqa/flake8) to permit the latest version. <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/PyCQA/flake8/commit/65a38c42a7f1a05ff8d99b313160754fc9b7a0d8"><code>65a38c4</code></a> Release 7.1.0</li> <li><a href="https://github.com/PyCQA/flake8/commit/34c97e046a459b0682c82660f16c620369abd6b7"><code>34c97e0</code></a> Merge pull request <a href="https://redirect.github.com/pycqa/flake8/issues/1939">#1939</a> from PyCQA/new-pycodestyle</li> <li><a href="https://github.com/PyCQA/flake8/commit/defd315175b7b77472affb61a410e5720dabdc1a"><code>defd315</code></a> latest pycodestyle</li> <li><a href="https://github.com/PyCQA/flake8/commit/408d4d695c71b0b232cea576876e757c87a3379c"><code>408d4d6</code></a> Merge pull request <a href="https://redirect.github.com/pycqa/flake8/issues/1930">#1930</a> from mzagol/patch-1</li> <li><a href="https://github.com/PyCQA/flake8/commit/866ad729c64eea359960a8ac4e3f1201104ee55c"><code>866ad72</code></a> Add --extend-exclude to the TOC</li> <li><a href="https://github.com/PyCQA/flake8/commit/33e508307ac4545a45472fdc32c6eaadbc7b9580"><code>33e5083</code></a> Merge pull request <a href="https://redirect.github.com/pycqa/flake8/issues/1923">#1923</a> from Viicos/entry-points-docs</li> <li><a href="https://github.com/PyCQA/flake8/commit/6659b213c9aa8fa49235e13a365fcd34f58cbc6b"><code>6659b21</code></a> Fix toctree ordering in index</li> <li><a href="https://github.com/PyCQA/flake8/commit/ba0f56610adbd4d8733772ce1c63efcab1b70079"><code>ba0f566</code></a> Use explicit external references</li> <li><a href="https://github.com/PyCQA/flake8/commit/350f2545fd3ec75640a1605e4995a2f921e8b38b"><code>350f254</code></a> Use explicit external references</li> <li><a href="https://github.com/PyCQA/flake8/commit/49f52a8598d8a934b07f367a1b3ad87dbe51be5b"><code>49f52a8</code></a> Update documentation regarding entry points</li> <li>Additional commits viewable in <a href="https://github.com/pycqa/flake8/compare/0.1...7.1.0">compare view</a></li> </ul> </details> <br /> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) </details> Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
### Details: - This PR extends the current `SwiGLU` primitive to support `GeGLU` that has Gelu activations instead of Swish. - This GeGLU patterns can be found in stable diffusion models. ### Tickets: - 143486
### Tickets: - *142219*
…t#25079) Checking if CVS-127101 still reproduces (it doesn't: https://github.com/openvinotoolkit/openvino/actions/runs/9550495926/job/26323459000?pr=25079, see passing MO unit tests)
Incorrect check resulted into decompostion of different RNN operations
### Details: - *Support `aten::col2im`* ### Tickets: - *CVS-101054*
Transformation that undoing make_stateful from optimum-intel.
### How to use in Python
```python
import openvino as ov
from openvino._offline_transformations import stateful_to_stateless_transformation
core = ov.Core()
model = core.read_model('your_chatty_stateful_model_right_from_vanilla_optimum_intel.xml')
stateful_to_stateless_transformation(model)
# use `model`
```
### How to use in C++
```c++
#include <openvino/openvino.hpp>
#include <openvino/pass/stateful_to_stateless.hpp>
int main() {
auto core = ov::Core();
auto model = core.read_model("your_chatty_stateful_model_right_from_vanilla_optimum_intel.xml");
ov::pass::StatefulToStateless().run_on_model(model);
// use `model`
}
```
### TODO
- [x] Restore the original order of inputs/output (now they are not
globally ordered, but kv inputs corresponds to kv outputs by indices
with a proper offset).
- [x] Restore the original names of inputs and outputs based on
optimum-intel conventions in make_stateful.
### Details: - Added path to tokenizers build to find ov wheel (openvinotoolkit/openvino_tokenizers#183) ### Tickets: - *ticket-id*
…envinotoolkit#25381) ### Details: - Add TensorOffset support key in adaptive pooling ref kernel as the kernel is using INPUT0_GET_INDEX() macro ### Tickets: - 144826
…d correct the number of submodels (openvinotoolkit#25224) ### Details: - Fix qwen1.5-14b-chat with HETERO pipeline parallelism Add supported to patten: ``` ReadValue->Gather->Concat |------>ShapeOf(fused on other different affinity node) ->.... ``` - Correct the value of HETERO_NUMBER_OF_SUBMODELS by subtracting the number of independent submodels to reduce confusion ### Tickets: - *ticket-id*
### Details: - Pass empty cache_dir to CPU plugin if cache_dir is set by user app when enable_startup_fallback or enable_runtime_fallback is true ### Tickets: - 141026 --------- Co-authored-by: Chen Peter <[email protected]>
### Details: - Remove unnecessary copies from Conditional Compilation an common util components. ### Tickets: - N/A
…c_if_needed (openvinotoolkit#25345) ### Details: - Fix layout count comparison by using actual byte value ### Tickets: - 144293
### Tickets: - *142826*
…openvinotoolkit#25377) ### Details: - This PR fixed the `convolution_gpu_bfyx_f16_1x1` kernel to set the vector size of fused ops to updated `blockWidth`.
…5344) ### Details: - *Disable avx2 asymmetric dynamic quantization* - *...* ### Tickets: - *ticket-id*
### Details: - Difference in boolean results was causing an unexpected issue while validation. We don't need to know max difference of a boolean array, just highlight it exists. ### Tickets: - 137495 --------- Co-authored-by: Evgenya Nugmanova <[email protected]>
### Details: Fix forked oneDNN JIT reorder kernel for the case of bf16 precision and AVX2 ISA. OneDNN fork PR: openvinotoolkit/oneDNN#253 ### Tickets: - CVS-145874
…penvinotoolkit#25395) This reverts commit 01dc53a. Reverting this PR because it is causing LLM failures ### Tickets: - 145784
### Details: - Added wstring support to save_model(). ### Tickets: - 144915
### Details: - Apply in place crop for dynamic shape at runtime ### Tickets: - 143096 --------- Signed-off-by: Andrew Park <[email protected]>
…otoolkit#25328) ### Details: - Use vectorization for conversion of bf16 -> f16, f32 ### Related PR: - openvinotoolkit#25248 ### Tickets: - CVS-145803 --------- Co-authored-by: Sergey Lyalin <[email protected]>
### Details: 2 issues have been fixed - ACL supports tensor rank up to 4 for both Reduce kernels - NEReduceMean and NEReductionOperation (https://github.com/ARM-software/ComputeLibrary/blob/505adb91d40e05b3f80a075a4467a78a253395e1/src/runtime/NEON/functions/NEReductionOperation.cpp#L78) The fix checks axis for both NEReduceMean and NEReductionOperation. - `axisCast` method is updated to support NDHWC to NCDHW and vice versa conversion. ### Tickets: - *ticket-id*
…indices of Gather (openvinotoolkit#25221) ### Details: - Added u64 support for the indices of Gather. ### Tickets: - openvinotoolkit#23545 ### Test is passed: > [==========] Running 1 test from 1 test suite. [----------] Global test environment set-up. [----------] 1 test from conformance_Gather/ReadIRTest [ RUN ] conformance_Gather/ReadIRTest.Inference/Op=Gather.8_Type=f32_Shape=dynamic_IR=1c727cc96123227a9fe6c3079a497fd64a04f273bff45b5ea56a3c0d577eca8e_Device=TEMPLATE_Config=() MEM_USAGE=31860KB [ CONFORMANCE ] Influence coefficient: 4.72987e-05 [ PLUGIN ] `SubgraphBaseTest::compile_model()` is started [ PLUGIN ] `SubgraphBaseTest::compile_model()` is finished successfully. Duration is 0.143964s RANGE FOR PARAMETER: Gather-8_0 start from: 0.000000 range: 8 resolution: 32 seed: 1 RANGE FOR PARAMETER: Gather-8_1 start from: 0.000000 range: 15 resolution: 1 seed: 1 [ PLUGIN ] `SubgraphBaseTest::get_plugin_outputs()` is started [ PLUGIN ] `SubgraphBaseTest::get_plugin_outputs()` is finished successfully. Duration is 0.0288465s [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is started [ REFERENCE ] Calculate reference in runtime [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is started [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is finished successfully. Duration is 0.0094897s [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is finished successfully. Duration is 0.0109919s [ COMPARATION ] `ov_tensor_utils.hpp::compare()` is started [ COMPARATION ] `ov_tensor_utils.hpp::compare()` is finished successfully. Duration is 0.0071786s RANGE FOR PARAMETER: Gather-8_0 start from: 0.000000 range: 8 resolution: 32 seed: 1 RANGE FOR PARAMETER: Gather-8_1 start from: 0.000000 range: 15 resolution: 1 seed: 1 [ PLUGIN ] `SubgraphBaseTest::get_plugin_outputs()` is started [ PLUGIN ] `SubgraphBaseTest::get_plugin_outputs()` is finished successfully. Duration is 0.0033599s [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is started [ REFERENCE ] Calculate reference in runtime [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is started [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is finished successfully. Duration is 0.0086416s [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is finished successfully. Duration is 0.0105932s [ COMPARATION ] `ov_tensor_utils.hpp::compare()` is started [ REFERENCE ] `SubgraphBaseTest::calculate_refs()` is finished successfully. Duration is 0.010214s [ COMPARATION ] `ov_tensor_utils.hpp::compare()` is started [ COMPARATION ] `ov_tensor_utils.hpp::compare()` is finished successfully. Duration is 0.0082425s [ OK ] conformance_Gather/ReadIRTest.Inference/Op=Gather.8_Type=f32_Shape=dynamic_IR=1c727cc96123227a9fe6c3079a497fd64a04f273bff45b5ea56a3c0d577eca8e_Device=TEMPLATE_Config=() (381 ms) [----------] 1 test from conformance_Gather/ReadIRTest (383 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test suite ran. (435 ms total) [ PASSED ] 1 test.
…lkit#25151) ### Details: - Removed workaround which caused unexpected collisions in case seed -1 < seed < 1 ### Tickets: - 123003
### Details: - Moved to VersionRange usage - Moved a translators registration near a translator code - Simplified file structure - Implemented support of *.lib build - Implemented a supported op customization by removing op-files from the build ### Tickets: - N/A
### Details: - Move instead of copy - Avoid throw from destructor ### Tickets: - CVS-145054
|
@ceciliapeng2011, are you sure? :) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.