-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[GPU]Move RMSFusion pass ahead of ConvertPrecision #172
[GPU]Move RMSFusion pass ahead of ConvertPrecision #172
Commits on Aug 28, 2024
-
Property for model cache encryption/decryption functions (openvinotoo…
…lkit#24417) ### Details: - *add new property for model cache encryption/decryption function* - *encrypt/decrypt topology in CPU model cache if the callbacks are provided.* ### Tickets: - *CVS-139600* --------- Co-authored-by: Chen Peter <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4d716ba - Browse repository at this point
Copy the full SHA 4d716baView commit details -
[CPU] Fix BatchToSpace size_t underflow for the second dimension (ope…
…nvinotoolkit#26236) size_t underflow might happen for the second dimension as well, for example when layout is ndhwc. New test cases extended for `ndhwc` layout as well ### Tickets: - 14877
Configuration menu - View commit details
-
Copy full SHA for 9c7ac64 - Browse repository at this point
Copy the full SHA 9c7ac64View commit details -
Expose explicit openvino::threading target for parallel.hpp backend (o…
…penvinotoolkit#26269) ### Details: - Currently, threading library like TBB is provided by openvino::runtime itself. It does not work well when multiple find_package(OpenVINO) are used within the project and with different `COMPONENTS` - E.g. here openvinotoolkit/openvino.genai#794 we have to add Threading component even for cases, when it's not really used, because find_package within nested subdirectory populates properties of openvino::runtime which is found in internal directory. - Example of usage openvinotoolkit/openvino_tokenizers#236
Configuration menu - View commit details
-
Copy full SHA for 18bb9cf - Browse repository at this point
Copy the full SHA 18bb9cfView commit details -
[DOCS] Fix for issues with alignment in Model Caching graphs for mast…
…er (openvinotoolkit#26273) Fixed alignment in model caching graphs.
Configuration menu - View commit details
-
Copy full SHA for 2e5189b - Browse repository at this point
Copy the full SHA 2e5189bView commit details -
[GPU] Fix default internal buffers sizes for SDPA (openvinotoolkit#26266
) ### Details: - This PR fixes incorrect default buffers sizes configuration for static model in sdpa_opt kernel ### Tickets: - [CVS-150773](https://jira.devtools.intel.com/browse/CVS-150773)
Configuration menu - View commit details
-
Copy full SHA for 382ac84 - Browse repository at this point
Copy the full SHA 382ac84View commit details -
NPUW: changed _mm_loadu_si64 avx instruction to _mm_loadl_epi64, to s…
…upport gcc < 9.0 (openvinotoolkit#26190) ### Details: - trying to compile npu plugin with gcc 8.5 or lower version resulted into issues with unsupported intrinsics ``` c++ <source>:6:17: error: '_mm_loadu_si64' was not declared in this scope __m128i a = _mm_loadu_si64((const __m128i *)(data)); ^~~~~~~~~~~~~~ <source>:6:17: note: suggested alternative: '_mm_loadl_epi64' __m128i a = _mm_loadu_si64((const __m128i *)(data)); ``` - technically they might be different since _mm_loadu_si64 specifically doesn't require data alignment while _mm_loadl_epi64 not specifying them - checked generated asm code(gcc 14.1) , well they are different, but tests with memory loading at 8 bits step showed no difference, so I assume both functions can work on unaligned arrays ![image](https://github.com/user-attachments/assets/c70ea4c2-2ab2-487d-b611-83ffe1a34ad3) - npuw unit test soon will be available in OV, with ongoing PR openvinotoolkit#25780, while i've tested them locally with new intrinsics ### Tickets: - *ticket-id* - 136269 Co-authored-by: Dmitry Matveev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e2bce04 - Browse repository at this point
Copy the full SHA e2bce04View commit details -
[GPU] out-of-bound access fix for permute_tile_8x8_4x4 (openvinotoolk…
…it#26267) ### Details: - Fix out-of-bound access from permute kernel - It happened because the kernel loaded with vload even at the boundary ### Tickets: - 150360
Configuration menu - View commit details
-
Copy full SHA for de65193 - Browse repository at this point
Copy the full SHA de65193View commit details -
[GPU] Do not apply crop optimization for squeeze if the crop axis mat…
…ches the squeeze axis (openvinotoolkit#26277) ### Details: - Do not apply crop optimization for squeeze if the crop axis matches the squeeze axis
Configuration menu - View commit details
-
Copy full SHA for 91c909e - Browse repository at this point
Copy the full SHA 91c909eView commit details -
[CI] [GHA] Fix
CCACHE_REMOTE_DIR
value if in merge group (openvinot……oolkit#26268) ### Details: - Apparently, `github.event.merge_group.base_ref` for merge group checks does not resolve into the same thing as `github.base_ref` for pull request checks. It is an issue since merge queue beta: [1](https://github.com/orgs/community/discussions/40277) - `github.event.merge_group.base_ref` resolves into `refs/heads/master` while we need only the `master` part to construct the remote cache directory. ### Tickets: - *150744*
Configuration menu - View commit details
-
Copy full SHA for 20afc50 - Browse repository at this point
Copy the full SHA 20afc50View commit details -
Fix crash caused by empty string (openvinotoolkit#26250)
### Details: - *item1* - *...* ### Tickets: - *ticket-id* --------- Signed-off-by: Chen Peter <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c22da4f - Browse repository at this point
Copy the full SHA c22da4fView commit details -
Configuration menu - View commit details
-
Copy full SHA for c4b2739 - Browse repository at this point
Copy the full SHA c4b2739View commit details -
[NPU] Change log level to debug (openvinotoolkit#26282)
### Details: - *Change log level to debug* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for c86004e - Browse repository at this point
Copy the full SHA c86004eView commit details -
Configuration menu - View commit details
-
Copy full SHA for 94173d8 - Browse repository at this point
Copy the full SHA 94173d8View commit details -
Don't call paddle module if it's not found (openvinotoolkit#26286)
### Details: - To avoid cmake warnings when PDPD is not installed
Configuration menu - View commit details
-
Copy full SHA for 44caa93 - Browse repository at this point
Copy the full SHA 44caa93View commit details -
[GPU] Fix data size in test (openvinotoolkit#26287)
### Details: - Fix data size in test ### Tickets: - [*CVS-148605*](https://jira.devtools.intel.com/browse/CVS-148605)
Configuration menu - View commit details
-
Copy full SHA for 55723af - Browse repository at this point
Copy the full SHA 55723afView commit details
Commits on Aug 29, 2024
-
[GPU] Fix the issue missing data padding info in update_shape (openvi…
…notoolkit#26276) ### Details: - *Do not set shape_changed to true when shape is not changed.* - *Fix output layout does not have updated data_padding which is calculated by calc_output_layouts in update_shape* ### Tickets: - *149773*
Configuration menu - View commit details
-
Copy full SHA for 46d1a4d - Browse repository at this point
Copy the full SHA 46d1a4dView commit details -
Performance fix for aot_autograd (openvinotoolkit#26139)
### Details: - Fix the performance issues in aot_autograd path where constants are being treated as inputs. ### Tickets: - https://jira.devtools.intel.com/browse/CVS-139183
Configuration menu - View commit details
-
Copy full SHA for 91744e1 - Browse repository at this point
Copy the full SHA 91744e1View commit details -
[Common] MarkDequantizationSubgraph: avoid modification of RT Info fo…
…r fold_multiply_const=true option (openvinotoolkit#26280) ### Details: - The PR restores MarkDequantizationSubgraph transformation behavior to the state before openvinotoolkit#25783. This is required to avoid compressed zero-points constant folding during model conversion. ### Tickets: - *[CVS-150686](https://jira.devtools.intel.com/browse/CVS-150686)*
Configuration menu - View commit details
-
Copy full SHA for 305dca0 - Browse repository at this point
Copy the full SHA 305dca0View commit details -
[JAX FE] Support concatenate, integer_pow, reduce ops (openvinotoolki…
…t#26288) **Details:** This PR finalizes and allows to infer JAX ViT model **Tickets:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 090cdd7 - Browse repository at this point
Copy the full SHA 090cdd7View commit details -
[GPU] Enable activation fusing for softmax_gpu_bf kernel (openvinotoo…
…lkit#26291) ### Details: - *Significantly speed up LogSoftmax operation* ### Tickets: - *[148550](https://jira.devtools.intel.com/browse/CVS-148550)*
Configuration menu - View commit details
-
Copy full SHA for bb33333 - Browse repository at this point
Copy the full SHA bb33333View commit details -
[GPU] Add prefix support for PagedAttention operation (openvinotoolki…
…t#26289) ### Details: - Add prefix support for PagedAttention operation via existing pa_sdpa_opt kernel by processing subsequence's tokens in sequential mode, one by one for - Moved responsibility for intermediate buffers reallocation from the sdpa_opt kernel to kv_cache_update kernel (they both use the same data, but now one of these buffers can be reused by the pa_sdpa_opt kernel, so everything was moved to one place)
Configuration menu - View commit details
-
Copy full SHA for 2c99533 - Browse repository at this point
Copy the full SHA 2c99533View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9fe6dd5 - Browse repository at this point
Copy the full SHA 9fe6dd5View commit details -
Zero backend and compiler adapter to share single L0 context (openvin…
…otoolkit#26016) ### Details: - share the same L0 context from backend to compiler - and perform zeDestroy(context) in backend only ### Tickets: - *ticket-id* --------- Co-authored-by: Xin Wang <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5f5f027 - Browse repository at this point
Copy the full SHA 5f5f027View commit details -
[GPU]: Bitwise shifts: Added support for uint16 and int16 data type (o…
…penvinotoolkit#26251) ### Details: - Bitwise shift operations have support now for int16 an uint16 datatype - added unit and functional test
Configuration menu - View commit details
-
Copy full SHA for e9eac15 - Browse repository at this point
Copy the full SHA e9eac15View commit details -
[GPU] Add dynamic quantize group size for clDNN Fully-connected (open…
…vinotoolkit#26231) ### Details: - *item1* - *...* ### Tickets: - CVS-148548 --------- Signed-off-by: Min, Byung-il <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c78fef6 - Browse repository at this point
Copy the full SHA c78fef6View commit details -
[CI] [GHA] Introduce Debian 10 ARM (openvinotoolkit#26156)
### Tickets: - *148717*
Configuration menu - View commit details
-
Copy full SHA for fcd1138 - Browse repository at this point
Copy the full SHA fcd1138View commit details -
[GPU] Use ngraph Transpose+MatMul fusion instead of custom code in la…
…yout optimizer (openvinotoolkit#25708) ### Details: - Allow Tranpose+Matmul+[Transpose] fusion for static shapes - Allow Transpose in `MoveEltwiseUpThroughDataMov` for specific Transpose -> Eltwise -> MatMul case - Change order of `MoveEltwiseUpThroughDataMov` and `ConvertMatMulToFullyConnected` to simplify callback - Removed custom code for similar fusion in layout optimizer and related debug knob
Configuration menu - View commit details
-
Copy full SHA for 1f9a6f8 - Browse repository at this point
Copy the full SHA 1f9a6f8View commit details -
[GPU] Add Validate pass call after IncreasePositionIdsPrecision to en…
…sure proper data type propagation (openvinotoolkit#26299) ### Details: This patch adds Validate pass call after IncreasePositionIdsPrecision to ensure proper data type propagation With this change the accuracy of llama-3-8b INT8 (and other LLMs probably) can be restored to expected level Before: ``` | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr| |--------|------:|------|-----:|---------------|---|-----:|---|------| |wikitext| 2|none | 0|bits_per_byte |↓ |0.6030|± |N/A | | | |none | 0|byte_perplexity|↓ |1.5189|± |N/A | | | |none | 0|word_perplexity|↓ |9.3472|± |N/A | ``` After: ``` | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr| |--------|------:|------|-----:|---------------|---|-----:|---|------| |wikitext| 2|none | 0|bits_per_byte |↓ |0.5351|± |N/A | | | |none | 0|byte_perplexity|↓ |1.4490|± |N/A | | | |none | 0|word_perplexity|↓ |7.2664|± |N/A | ``` ### Tickets: - [CVS-147653](https://jira.devtools.intel.com/browse/CVS-147653)
Configuration menu - View commit details
-
Copy full SHA for 2cef478 - Browse repository at this point
Copy the full SHA 2cef478View commit details -
[GPU] Use recursive_mutex instead of mutex in compilation context (op…
…envinotoolkit#26295) ### Details: - Use recursive_mutex instead of mutex in compilation context. Because deadlock happens in context compilation(push_tash() -> remove_keys() in single thread) when CPU_PINNING=ON ### Tickets: - 150220
Configuration menu - View commit details
-
Copy full SHA for 79b7316 - Browse repository at this point
Copy the full SHA 79b7316View commit details -
[PT FE] Fix rt_info for decoder type (openvinotoolkit#26300)
### Details: - *Fix issue with not set decoder type rt_info* - *Remove frontend rt_info from model* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for b2369a9 - Browse repository at this point
Copy the full SHA b2369a9View commit details -
[GPU] Fix Interpolate 11 axes reading (openvinotoolkit#26260)
### Details: - Fix interpolate 11 op case when sizes or scales are not constant. The additional argument shadowed TensorAccessor underlying tensor on port 2 and therefore shape inference was reading wrong values ### Tickets: - 146022
Configuration menu - View commit details
-
Copy full SHA for 7802144 - Browse repository at this point
Copy the full SHA 7802144View commit details -
[CVS-149644] Fixed ONNX FE debug build (openvinotoolkit#26301)
### Details: - Fixed ONNX FE debug build - ONNX proto is base library and must be compiled with newer standard, otherwise onnx and onnx_proto are compiled with different C++ standard. ### Tickets: - CVS-149644
Configuration menu - View commit details
-
Copy full SHA for 98188ad - Browse repository at this point
Copy the full SHA 98188adView commit details -
Fix Sanitizer after TBB update (openvinotoolkit#26278)
### Details: - Introduce an environment variable to make TBB compatible with Sanitizer ### Tickets: - 150004
Configuration menu - View commit details
-
Copy full SHA for 3a2626a - Browse repository at this point
Copy the full SHA 3a2626aView commit details -
Update flake8 requirement from <=7.1.0 to <=7.1.1 in /src/bindings/py…
…thon (openvinotoolkit#26245) 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/cf1542cefa3e766670b2066dd75c4571d682a649"><code>cf1542c</code></a> Release 7.1.1</li> <li><a href="https://github.com/PyCQA/flake8/commit/939ea3d8d9d5d7d9f348420036af52df74f5ca09"><code>939ea3d</code></a> Merge pull request <a href="https://redirect.github.com/pycqa/flake8/issues/1949">#1949</a> from stephenfin/issue-1948</li> <li><a href="https://github.com/PyCQA/flake8/commit/bdcd5c2c0afadaf7c92a4b26d96055cecdd38cf3"><code>bdcd5c2</code></a> Handle escaped braces in f-strings</li> <li><a href="https://github.com/PyCQA/flake8/commit/2a811cc4d2aaed3e8eb5a9f04f08ccc8af7c0791"><code>2a811cc</code></a> Merge pull request <a href="https://redirect.github.com/pycqa/flake8/issues/1946">#1946</a> from Viicos/patch-1</li> <li><a href="https://github.com/PyCQA/flake8/commit/10314ad9e5236f1ddf70cb25c2854c93c0840b66"><code>10314ad</code></a> Fix wording of plugins documentation</li> <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>Additional commits viewable in <a href="https://github.com/pycqa/flake8/compare/0.1...7.1.1">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>
Configuration menu - View commit details
-
Copy full SHA for 10d1e4e - Browse repository at this point
Copy the full SHA 10d1e4eView commit details -
[GPU] unfusing invalid eltwise fused onednn fc layers for dynamic mod…
…els (openvinotoolkit#26284) ### Details: - In some cases, OneDNN fc layers requires fully broadcasted data for fused eltwise layers. - For dynamic models, it's not easy to know the layouts of fc and eltwise layers in model compilation time. - So, we need to execute unfused subgraphs for such cases. ### Tickets: - 146778
Configuration menu - View commit details
-
Copy full SHA for 13a6f31 - Browse repository at this point
Copy the full SHA 13a6f31View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9a60a62 - Browse repository at this point
Copy the full SHA 9a60a62View commit details -
[CI] [GHA] Introduce Android x64 workflow (openvinotoolkit#26086)
### Tickets: - *148716*
Configuration menu - View commit details
-
Copy full SHA for 3ffb013 - Browse repository at this point
Copy the full SHA 3ffb013View commit details -
[GHA] Don't set CI_BUILD_DEV_TAG on release branches (openvinotoolkit…
…#26313) Same as openvinotoolkit#26312 but for master
Configuration menu - View commit details
-
Copy full SHA for bb09495 - Browse repository at this point
Copy the full SHA bb09495View commit details
Commits on Aug 30, 2024
-
[CPU] Fixed invalid debug assert for weights compression path (openvi…
…notoolkit#26296) ### Details: - Some weights compression optimizations implemented inside oneDNN fork resulted in invalide state of particular debug assertion logic. oneDNN fork PR: openvinotoolkit/oneDNN#263
Configuration menu - View commit details
-
Copy full SHA for 31b5b4c - Browse repository at this point
Copy the full SHA 31b5b4cView commit details -
[DOCS] pypi publishing page adjustment (openvinotoolkit#25746)
A proposition of changes to the pypi page
Configuration menu - View commit details
-
Copy full SHA for 5489b36 - Browse repository at this point
Copy the full SHA 5489b36View commit details -
Bump actions/setup-python from 5.1.1 to 5.2.0 (openvinotoolkit#26318)
Bumps [actions/setup-python](https://github.com/actions/setup-python) from 5.1.1 to 5.2.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/actions/setup-python/releases">actions/setup-python's releases</a>.</em></p> <blockquote> <h2>v5.2.0</h2> <h2>What's Changed</h2> <h3>Bug fixes:</h3> <ul> <li>Add <code>.zip</code> extension to Windows package downloads for <code>Expand-Archive</code> Compatibility by <a href="https://github.com/priyagupta108"><code>@priyagupta108</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/916">actions/setup-python#916</a> This addresses compatibility issues on Windows self-hosted runners by ensuring that the filenames for Python and PyPy package downloads explicitly include the .zip extension, allowing the Expand-Archive command to function correctly.</li> <li>Add arch to cache key by <a href="https://github.com/Zxilly"><code>@Zxilly</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/896">actions/setup-python#896</a> This addresses issues with caching by adding the architecture (arch) to the cache key, ensuring that cache keys are accurate to prevent conflicts</li> </ul> <h3>Documentation changes:</h3> <ul> <li>Fix display of emojis in contributors doc by <a href="https://github.com/sciencewhiz"><code>@sciencewhiz</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/899">actions/setup-python#899</a></li> <li>Documentation update for caching poetry dependencies by <a href="https://github.com/gowridurgad"><code>@gowridurgad</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/908">actions/setup-python#908</a></li> </ul> <h3>Dependency updates:</h3> <ul> <li>Bump <code>@iarna/toml</code> version from 2.2.5 to 3.0.0 by <a href="https://github.com/priya-kinthali"><code>@priya-kinthali</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/912">actions/setup-python#912</a></li> <li>Bump pyinstaller from 3.6 to 5.13.1 by <a href="https://github.com/aparnajyothi-y"><code>@aparnajyothi-y</code></a> in <a href="https://redirect.github.com/actions/setup-python/pull/923">actions/setup-python#923</a></li> </ul> <h2>New Contributors</h2> <ul> <li><a href="https://github.com/sciencewhiz"><code>@sciencewhiz</code></a> made their first contribution in <a href="https://redirect.github.com/actions/setup-python/pull/899">actions/setup-python#899</a></li> <li><a href="https://github.com/priyagupta108"><code>@priyagupta108</code></a> made their first contribution in <a href="https://redirect.github.com/actions/setup-python/pull/916">actions/setup-python#916</a></li> <li><a href="https://github.com/Zxilly"><code>@Zxilly</code></a> made their first contribution in <a href="https://redirect.github.com/actions/setup-python/pull/896">actions/setup-python#896</a></li> <li><a href="https://github.com/aparnajyothi-y"><code>@aparnajyothi-y</code></a> made their first contribution in <a href="https://redirect.github.com/actions/setup-python/pull/923">actions/setup-python#923</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/setup-python/compare/v5...v5.2.0">https://github.com/actions/setup-python/compare/v5...v5.2.0</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/actions/setup-python/commit/f677139bbe7f9c59b41e40162b753c062f5d49a3"><code>f677139</code></a> Bump pyinstaller from 3.6 to 5.13.1 in /<strong>tests</strong>/data (<a href="https://redirect.github.com/actions/setup-python/issues/923">#923</a>)</li> <li><a href="https://github.com/actions/setup-python/commit/2bd53f9a4d1dd1cd21eaffcc01a7b91a8e73ea4c"><code>2bd53f9</code></a> Documentation update for caching poetry dependencies (<a href="https://redirect.github.com/actions/setup-python/issues/908">#908</a>)</li> <li><a href="https://github.com/actions/setup-python/commit/80b49d3ed89312896dbdcbefc2ddb159c7f8ca43"><code>80b49d3</code></a> fix: add arch to cache key (<a href="https://redirect.github.com/actions/setup-python/issues/896">#896</a>)</li> <li><a href="https://github.com/actions/setup-python/commit/036a5236741fd24c89eea80d1b76179e8e5f9214"><code>036a523</code></a> Fix: Add <code>.zip</code> extension to Windows package downloads for <code>Expand-Archive</code> C...</li> <li><a href="https://github.com/actions/setup-python/commit/04c1311429f7be71707d8ab66c7af8a14e54b938"><code>04c1311</code></a> Fix display of emojis in contributors doc (<a href="https://redirect.github.com/actions/setup-python/issues/899">#899</a>)</li> <li><a href="https://github.com/actions/setup-python/commit/cb6845644151e35f879e10f2f0896c3c8bee372c"><code>cb68456</code></a> Updated <code>@iarna/toml</code> version to 3.0.0 (<a href="https://redirect.github.com/actions/setup-python/issues/912">#912</a>)</li> <li>See full diff in <a href="https://github.com/actions/setup-python/compare/39cd14951b08e74b54015e9e001cdefcf80e669f...f677139bbe7f9c59b41e40162b753c062f5d49a3">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/setup-python&package-manager=github_actions&previous-version=5.1.1&new-version=5.2.0)](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>
Configuration menu - View commit details
-
Copy full SHA for ad81cbe - Browse repository at this point
Copy the full SHA ad81cbeView commit details -
[CPU] Add script for benchmark average counters aggregation (openvino…
…toolkit#25652) Example output: | Node type | Count | Sum (ms) | % | |:-------------------------------------|--------:|-----------:|-------:| | FullyConnected | 197 | 77.812 | 91.77 | | ScaledDotProductAttentionWithKVCache | 28 | 3.7 | 4.36 | | RoPE | 56 | 0.687 | 0.81 | | Reorder | 161 | 0.473 | 0.56 | | Multiply | 31 | 0.443 | 0.52 | | Transpose | 56 | 0.431 | 0.51 | | PowerStatic | 115 | 0.313 | 0.37 | | Add | 59 | 0.227 | 0.27 | | ReduceMean | 57 | 0.15 | 0.18 | | ReadValue | 56 | 0.134 | 0.16 | | Reshape | 200 | 0.114 | 0.13 | | Broadcast | 6 | 0.092 | 0.11 | | Convert | 8 | 0.092 | 0.11 | | Ceiling | 2 | 0.029 | 0.03 | | Select | 2 | 0.022 | 0.03 | | Concat | 6 | 0.019 | 0.02 | | Gather | 6 | 0.015 | 0.02 | | Unsqueeze | 2 | 0.01 | 0.01 | | Clamp | 2 | 0.009 | 0.01 | | Range | 1 | 0.009 | 0.01 | | Less | 1 | 0.005 | 0.01 | | Abs | 1 | 0.004 | 0 | | ShapeOf | 3 | 0.004 | 0 | | Constant | 745 | 0 | 0 | | Assign | 56 | 0 | 0 | | Parameter | 4 | 0 | 0 | | Result | 1 | 0 | 0 | | Total | 1862 | 84.794 | 100 |
Configuration menu - View commit details
-
Copy full SHA for d622afe - Browse repository at this point
Copy the full SHA d622afeView commit details -
[CPU][REFACTORING] Align inference pipeline for all scenarios (openvi…
…notoolkit#26211) ### Details: - Execute node with exception handling in all scenarios - Move all the profiling macros under ExecuteNode - Get numaId outside the main loop - Few minor changes
Configuration menu - View commit details
-
Copy full SHA for 50896e9 - Browse repository at this point
Copy the full SHA 50896e9View commit details -
[GPU] Sycl op example, base class cleanup, docs update (openvinotoolk…
…it#26304) ### Details: - Added naive FC sycl impl as an example and it will also be used to verify base class compilation in CI util we have any production sycl op. - Moved sycl readme to common gpu docs folder with some minor updates
Configuration menu - View commit details
-
Copy full SHA for fd25ea4 - Browse repository at this point
Copy the full SHA fd25ea4View commit details -
[Core] Validate model before serialization (openvinotoolkit#26184)
### Details: - Added model validate model before serialization. - Added test. ### Tickets: - CVS-133613
Configuration menu - View commit details
-
Copy full SHA for 01ddcd2 - Browse repository at this point
Copy the full SHA 01ddcd2View commit details -
[DOCS] Update notebooks section for master (openvinotoolkit#26331)
Updating notebook section with new content.
Configuration menu - View commit details
-
Copy full SHA for 6fef4a0 - Browse repository at this point
Copy the full SHA 6fef4a0View commit details -
[GPU] dump verbose log to file (openvinotoolkit#26244)
### Details: - New debug config for dumping verbose log into file
Configuration menu - View commit details
-
Copy full SHA for 446d2e2 - Browse repository at this point
Copy the full SHA 446d2e2View commit details -
Update setuptools requirement from <73.1.0,>=65.6.1 to >=65.6.1,<74.1…
….0 in /src/bindings/python (openvinotoolkit#26322) Updates the requirements on [setuptools](https://github.com/pypa/setuptools) to permit the latest version. <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/pypa/setuptools/blob/main/NEWS.rst">setuptools's changelog</a>.</em></p> <blockquote> <h1>v74.0.0</h1> <h2>Features</h2> <ul> <li>Changed the type of error raised by <code>setuptools.command.easy_install.CommandSpec.from_param</code> on unsupported argument from <code>AttributeError</code> to <code>TypeError</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4548">#4548</a>)</li> <li>Added detection of ARM64 variant of MSVC -- by :user:<code>saschanaz</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4553">#4553</a>)</li> <li>Made <code>setuptools.package_index.Credential</code> a <code>typing.NamedTuple</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4585">#4585</a>)</li> <li>Reraise error from <code>setuptools.command.easy_install.auto_chmod</code> instead of nonsensical <code>TypeError: 'Exception' object is not subscriptable</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4593">#4593</a>)</li> <li>Fully typed all collection attributes in <code>pkg_resources</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4598">#4598</a>)</li> <li>Automatically exclude <code>.tox|.nox|.venv</code> directories from <code>sdist</code>. (<a href="https://redirect.github.com/pypa/setuptools/issues/4603">#4603</a>)</li> </ul> <h2>Deprecations and Removals</h2> <ul> <li>Removed the monkeypatching of distutils._msvccompiler. Now all compiler logic is consolidated in distutils. (<a href="https://redirect.github.com/pypa/setuptools/issues/4600">#4600</a>)</li> <li>Synced with pypa/distutils@58fe058e4, including consolidating Visual Studio 2017 support (<a href="https://redirect.github.com/pypa/setuptools/issues/4600">#4600</a>, <a href="https://redirect.github.com/pypa/distutils/issues/289">pypa/distutils#289</a><code>pypa/distutils#287</code><a href="https://redirect.github.com/pypa/setuptools/issues/4606">#4606</a>)</li> </ul> <h2>Misc</h2> <ul> <li><a href="https://redirect.github.com/pypa/setuptools/issues/4592">#4592</a></li> </ul> <h1>v73.0.1</h1> <h2>Bugfixes</h2> <ul> <li>Remove <code>abc.ABCMeta</code> metaclass from abstract classes. <code>pypa/setuptools#4503 <https://github.com/pypa/setuptools/pull/4503></code>_ had an unintended consequence of causing potential <code>TypeError: metaclass conflict: the metaclass of a derived class must be a (non-strict) subclass of the metaclasses of all its bases</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4579">#4579</a>)</li> </ul> <h1>v73.0.0</h1> <h2>Features</h2> <ul> <li>Mark abstract base classes and methods with <code>abc.ABC</code> and <code>abc.abstractmethod</code> -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4503">#4503</a>)</li> <li>Changed the order of type checks in <code>setuptools.command.easy_install.CommandSpec.from_param</code> to support any <code>collections.abc.Iterable</code> of <code>str</code> param -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4505">#4505</a>)</li> </ul> <h2>Bugfixes</h2> <ul> <li>Prevent an error in <code>bdist_wheel</code> if <code>compression</code> is set to a <code>str</code> (even if valid) after finalizing options but before running the command. -- by :user:<code>Avasam</code> (<a href="https://redirect.github.com/pypa/setuptools/issues/4383">#4383</a>)</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/pypa/setuptools/commit/98ad794354efecf4ed1f629d4e5f02feae00d2ae"><code>98ad794</code></a> Bump version: 73.0.1 → 74.0.0</li> <li><a href="https://github.com/pypa/setuptools/commit/b4fb91796ba9f9473280a69a3c8213066e5bc107"><code>b4fb917</code></a> Merge pull request <a href="https://redirect.github.com/pypa/setuptools/issues/4600">#4600</a> from pypa/debt/msvc-monkey</li> <li><a href="https://github.com/pypa/setuptools/commit/18a44d8f5660df9e23ee823a073b2d3238bc8293"><code>18a44d8</code></a> Add news fragment.</li> <li><a href="https://github.com/pypa/setuptools/commit/5f8215d9888cb555d2a206635f0f59421cc4afa9"><code>5f8215d</code></a> Merge pull request <a href="https://redirect.github.com/pypa/setuptools/issues/4548">#4548</a> from Avasam/from_param-TypeError</li> <li><a href="https://github.com/pypa/setuptools/commit/6928048a3c52370363fa47e7c2b9496ff0de0f79"><code>6928048</code></a> Merge branch 'main' into debt/msvc-monkey</li> <li><a href="https://github.com/pypa/setuptools/commit/11a6b596ed1453407061b3e57da04bd49c0adb91"><code>11a6b59</code></a> Merge pull request <a href="https://redirect.github.com/pypa/setuptools/issues/4606">#4606</a> from pypa/distutils-58fe058e4</li> <li><a href="https://github.com/pypa/setuptools/commit/903604bb3a648a26ce268753f6f05ce049336e5c"><code>903604b</code></a> Reraise sensible errors from auto_chmod (<a href="https://redirect.github.com/pypa/setuptools/issues/4593">#4593</a>)</li> <li><a href="https://github.com/pypa/setuptools/commit/8ec5b5aeef7a6cd80ca8c3291ad66acc3986069b"><code>8ec5b5a</code></a> Add missing news fragment for PR 4603</li> <li><a href="https://github.com/pypa/setuptools/commit/e90dfd568eef9dde8aa69a8a0ec1a7e692c532c8"><code>e90dfd5</code></a> Exclude top-level <code>.tox|.nox|.venv</code> from sdist (<a href="https://redirect.github.com/pypa/setuptools/issues/4603">#4603</a>)</li> <li><a href="https://github.com/pypa/setuptools/commit/ef2957a952244e4dfd179d29e1eaaa2cd83a1e26"><code>ef2957a</code></a> Reraise sensible errors from auto_chmod</li> <li>Additional commits viewable in <a href="https://github.com/pypa/setuptools/compare/v65.6.1...v74.0.0">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>
Configuration menu - View commit details
-
Copy full SHA for 19eb02f - Browse repository at this point
Copy the full SHA 19eb02fView commit details -
[TF FE] Stabilize TF2 Keras layer test for StackedRNNCells (openvinot…
…oolkit#26329) **Details:** Stabilize TF2 Keras layer test for StackedRNNCells **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for aa4d550 - Browse repository at this point
Copy the full SHA aa4d550View commit details -
[GHA] Increase timeout for PyTorch Models tests in precommit (openvin…
…otoolkit#26341) ### Tickets: - 139371
Configuration menu - View commit details
-
Copy full SHA for 6d31a50 - Browse repository at this point
Copy the full SHA 6d31a50View commit details -
Disabled LTO for gcc older than 9.0 (openvinotoolkit#26330)
### Details: - Gcc 8.x has multiple bug for LTO - Let's have 9.x and higher as required for LTO ### Tickets: - CVS-150950
Configuration menu - View commit details
-
Copy full SHA for 04e0fea - Browse repository at this point
Copy the full SHA 04e0feaView commit details
Commits on Sep 1, 2024
-
[Good First Issue][TF FE]: Support complex tensors for Pack operation (…
…openvinotoolkit#25193) ### Details: - ***Addresses the issue** : [[Good First Issue][TF FE]: Support complex tensors for Pack operation openvinotoolkit#22954](openvinotoolkit#22954 - *Added support for complex tensors for pack operation* - *Let me know if any changes are required* --------- Co-authored-by: Michal Lukaszewski <[email protected]> Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 3983c35 - Browse repository at this point
Copy the full SHA 3983c35View commit details -
[TF FE]: Support complex tensors for Sub operation (openvinotoolkit#2…
…6342) ### Details: - Support complex tensors for Sub operation + tests ### Tickets: - [None](openvinotoolkit#22948)
Configuration menu - View commit details
-
Copy full SHA for a6d92e3 - Browse repository at this point
Copy the full SHA a6d92e3View commit details -
Bump actions/upload-artifact from 4.3.4 to 4.4.0 (openvinotoolkit#26343)
Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4.3.4 to 4.4.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/actions/upload-artifact/releases">actions/upload-artifact's releases</a>.</em></p> <blockquote> <h2>v4.4.0</h2> <h2>Notice: Breaking Changes
⚠️ </h2> <p>We will no longer include hidden files and folders by default in the <code>upload-artifact</code> action of this version. This reduces the risk that credentials are accidentally uploaded into artifacts. Customers who need to continue to upload these files can use a new option, <code>include-hidden-files</code>, to continue to do so.</p> <p>See <a href="https://github.blog/changelog/2024-08-19-notice-of-upcoming-deprecations-and-breaking-changes-in-github-actions-runners/">"Notice of upcoming deprecations and breaking changes in GitHub Actions runners"</a> changelog and <a href="https://redirect.github.com/actions/upload-artifact/issues/602">this issue</a> for more details.</p> <h2>What's Changed</h2> <ul> <li>Exclude hidden files by default by <a href="https://github.com/joshmgross"><code>@joshmgross</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/598">actions/upload-artifact#598</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4.3.6...v4.4.0">https://github.com/actions/upload-artifact/compare/v4.3.6...v4.4.0</a></p> <h2>v4.3.6</h2> <h2>What's Changed</h2> <ul> <li>Revert to <code>@actions/artifact</code> 2.1.8 by <a href="https://github.com/robherley"><code>@robherley</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/594">actions/upload-artifact#594</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4...v4.3.6">https://github.com/actions/upload-artifact/compare/v4...v4.3.6</a></p> <h2>v4.3.5</h2> <h2>What's Changed</h2> <ul> <li>Bump <code>@actions/artifact</code> to v2.1.9 by <a href="https://github.com/robherley"><code>@robherley</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/588">actions/upload-artifact#588</a> <ul> <li>Fixed artifact upload chunk timeout logic <a href="https://redirect.github.com/actions/toolkit/pull/1774">#1774</a></li> <li>Use lazy stream to prevent issues with open file limits <a href="https://redirect.github.com/actions/toolkit/pull/1771">#1771</a></li> </ul> </li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4.3.4...v4.3.5">https://github.com/actions/upload-artifact/compare/v4.3.4...v4.3.5</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/actions/upload-artifact/commit/50769540e7f4bd5e21e526ee35c689e35e0d6874"><code>5076954</code></a> Merge pull request <a href="https://redirect.github.com/actions/upload-artifact/issues/598">#598</a> from actions/joshmgross/exclude-hidden-files</li> <li><a href="https://github.com/actions/upload-artifact/commit/d52396ac5d312b1bda1b9fe2ae55d862c65abd68"><code>d52396a</code></a> Add a warning about enabling <code>include-hidden-files</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/710f36207581d49e465ccbed3e007c317290fe11"><code>710f362</code></a> Remove "merged" from <code>include-hidden-files</code> input description</li> <li><a href="https://github.com/actions/upload-artifact/commit/3b315f26f6f828f74089e1863e67e60e48e37563"><code>3b315f2</code></a> <code>npm run release</code> again 🙂</li> <li><a href="https://github.com/actions/upload-artifact/commit/3be2180eb79b56341b1b127ca06a5adba2a5a3e4"><code>3be2180</code></a> Remove another trailing comma</li> <li><a href="https://github.com/actions/upload-artifact/commit/453e8d0a40444939146df5febed0b1221b9dd073"><code>453e8d0</code></a> Update glob license</li> <li><a href="https://github.com/actions/upload-artifact/commit/0a398c14804ead69f18a166f0a99f91239261ee5"><code>0a398c1</code></a> <code>npm run release</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/a0c40cf60283f329afac2bc0fa77f1a59fa11e6e"><code>a0c40cf</code></a> Update to latest <code>@actions/glob</code> and fix tests</li> <li><a href="https://github.com/actions/upload-artifact/commit/acb59e47763b6e601a344d04cf3c082ee336b709"><code>acb59e4</code></a> <code>lint</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/cb6558bb10fe4afe4054d0be4b3136e673eb5e7f"><code>cb6558b</code></a> Exclude hidden files by default</li> <li>Additional commits viewable in <a href="https://github.com/actions/upload-artifact/compare/0b2256b8c012f0828dc542b3febcab082c67f72b...50769540e7f4bd5e21e526ee35c689e35e0d6874">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/upload-artifact&package-manager=github_actions&previous-version=4.3.4&new-version=4.4.0)](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>Configuration menu - View commit details
-
Copy full SHA for c3179ee - Browse repository at this point
Copy the full SHA c3179eeView commit details -
Bump opencv-python from 4.6.0.66 to 4.8.1.78 in /tests/e2e_tests (ope…
…nvinotoolkit#26344) Bumps [opencv-python](https://github.com/opencv/opencv-python) from 4.6.0.66 to 4.8.1.78. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/opencv/opencv-python/releases">opencv-python's releases</a>.</em></p> <blockquote> <h2>4.8.1.78</h2> <p>OpenCV 4.8.1 release.</p> <p>Important changes:</p> <ul> <li>WebP security update for <a href="https://nvd.nist.gov/vuln/detail/CVE-2023-4863">CVE-2023-4863</a></li> </ul> <h2>4.8.0.76</h2> <p>Adds cv2.typing to package. Close <a href="https://redirect.github.com/opencv/opencv-python/issues/869">#869</a></p> <h2>4.8.0.74</h2> <p>Important changes:</p> <ul> <li><a href="https://redirect.github.com/opencv/opencv/pull/20370">#20370</a> Python typing stubs.</li> <li><a href="https://redirect.github.com/opencv/opencv/pull/23350">#23350</a> Fix reference counting errors in registerNewType.</li> <li><a href="https://redirect.github.com/opencv/opencv/pull/23399">#23399</a>, <a href="https://redirect.github.com/opencv/opencv/pull/23436">#23436</a>, <a href="https://redirect.github.com/opencv/opencv/pull/23138">#23138</a> Fixed ChAruco and diamond boards detector bindings.</li> <li><a href="https://redirect.github.com/opencv/opencv/pull/23371">#23371</a> Added bindings to allow GpuMat and Stream objects to be initialized from memory initialized in other libraries</li> <li><a href="https://redirect.github.com/opencv/opencv/pull/23691">#23691</a> np.float16 support.</li> <li>Python bindings for RotatedRect, CV_MAKETYPE, CV_8UC(n).</li> <li>Several build fixes for OpenCV-Python package</li> </ul> <h2>4.7.0.72</h2> <p>OpenCV 4.7.0 with various distribution bug fixes.</p> <ul> <li>Mac OS 11 support.</li> <li>Old Linux support with zlib version older than 1.9.</li> <li>Package build fixes for Python 11 on Musl C based system (Alpine).</li> </ul> <h2>4.7.0.70</h2> <p>OpenCV 4.7.0 with various distribution bug fixes.</p> <ul> <li>Mac OS 11 support.</li> <li>Old Linux support with zlib version older than 1.9.</li> <li>Package build fixes for Python 11 on Musl C based system (Alpine).</li> </ul> <h2>4.7.0.68</h2> <p>opencv-python: <a href="https://pypi.org/project/opencv-python/">https://pypi.org/project/opencv-python/</a> opencv-contrib-python: <a href="https://pypi.org/project/opencv-contrib-python/">https://pypi.org/project/opencv-contrib-python/</a> opencv-python-headless: <a href="https://pypi.org/project/opencv-python-headless/">https://pypi.org/project/opencv-python-headless/</a> opencv-contrib-python-headless: <a href="https://pypi.org/project/opencv-contrib-python-headless/">https://pypi.org/project/opencv-contrib-python-headless/</a></p> <p><a href="https://github.com/opencv/opencv/releases/tag/4.7.0">OpenCV 4.7.0</a></p> <p>Changes:</p> <ul> <li>Updated third-party libraries to fix potential vulnerabilities.</li> <li>Dropped Python 3.6 support.</li> <li>Added Python 3.11 support.</li> </ul> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li>See full diff in <a href="https://github.com/opencv/opencv-python/commits">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=opencv-python&package-manager=pip&previous-version=4.6.0.66&new-version=4.8.1.78)](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) You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/openvinotoolkit/openvino/network/alerts). </details> Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for e581a0f - Browse repository at this point
Copy the full SHA e581a0fView commit details
Commits on Sep 2, 2024
-
[GPU] Do not preallocate memory if desired size reaches the maximum s…
…upported buffer size (openvinotoolkit#26340) ### Details: - This patch fixes an issue where requested buffer meets the requirements of the maximum buffer allocation size, but applying preallocation causes the limits to be reached, resulting in an OpenCL error
Configuration menu - View commit details
-
Copy full SHA for 0d070e4 - Browse repository at this point
Copy the full SHA 0d070e4View commit details -
Fix CS issues (openvinotoolkit#26317)
### Details: - Fix CS issues - *...* ### Tickets: - *ticket-id* Signed-off-by: Zhai, Xuejun <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8168d1a - Browse repository at this point
Copy the full SHA 8168d1aView commit details -
[GPU] Small refactoring & cleanup (openvinotoolkit#26349)
### Details: - `aggregate_events` method is now a part of stream to avoid duplication in different places - Removed unnecessary prior_box primitive registration which called wait_for_events impl - Removed unnecessary `program_node::generates_dynamic_output` method - Removed unused `form_tags_by_ndims` map - Removed unused onednn eltwise registration - Minor fixes for broadcast and non zero layout inference - Minor cleanup
Configuration menu - View commit details
-
Copy full SHA for a4621ac - Browse repository at this point
Copy the full SHA a4621acView commit details -
[GPU] Add reqd_work_group_size attribute to softmax bf kernel (openvi…
…notoolkit#26345) ### Details: - Add reqd_work_group_size attribute to softmax bf kernel - Slight performance improvement ### Tickets: - 151211
Configuration menu - View commit details
-
Copy full SHA for 810efc9 - Browse repository at this point
Copy the full SHA 810efc9View commit details -
[GPU]: Added support for bitwise_and, bitwise_or and bitwise_xor (ope…
…nvinotoolkit#26325) ### Details: - Added basic support for bitwise or, xor, and - Add proper unit and functional tests ### Tickets: - CVS-148540
Configuration menu - View commit details
-
Copy full SHA for 1ceff40 - Browse repository at this point
Copy the full SHA 1ceff40View commit details -
[DOCS] Fixing snippets in Obtaining a Stateful OpenVINO Model (openvi…
…notoolkit#26355) Removing duplicate target names from `csv-table`'s in GenAI performance benchmarks. Fixing issues related to missing Python script file in snippets, in the `Obtaining a Stateful OpenVINO Model` article.
Configuration menu - View commit details
-
Copy full SHA for 62923a2 - Browse repository at this point
Copy the full SHA 62923a2View commit details -
[PT FE] Set target name as PythonOp name (openvinotoolkit#26327)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for e0c7c40 - Browse repository at this point
Copy the full SHA e0c7c40View commit details -
[PT FE] Enable layer tests on Windows (openvinotoolkit#26332)
### Details: - *Enable layer tests on Windows* - *Continuation of openvinotoolkit#23984* ### Tickets: - *CVS-115085*
Configuration menu - View commit details
-
Copy full SHA for d74a392 - Browse repository at this point
Copy the full SHA d74a392View commit details -
[TF FE] Support TF 2.17 and stick exact TF version for validation (op…
…envinotoolkit#26014) **Details:** Stick exact version for TF FE validation. This approach is more transparent about what TF version is exactly tested against. We will select the highest available version for each platform. **Ticket:** 150601 --------- Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6ff1420 - Browse repository at this point
Copy the full SHA 6ff1420View commit details
Commits on Sep 3, 2024
-
[OV JS] Use released openvino-node package in samples (openvinotoolki…
…t#26270) ### Details: - Replace `openvino-node` package in samples to do not build openvino to run js samples ### Tickets: - 149668
Configuration menu - View commit details
-
Copy full SHA for 27709d8 - Browse repository at this point
Copy the full SHA 27709d8View commit details -
[GPU] Layout optimizer as program member (openvinotoolkit#26359)
### Details: - Layout optimizer is made a member of program class to be able to use it during implementation selection
Configuration menu - View commit details
-
Copy full SHA for 0bdf5d5 - Browse repository at this point
Copy the full SHA 0bdf5d5View commit details -
[TF FE] Stabilize layer tests for Adjust operations on all platforms (o…
…penvinotoolkit#26370) **Details:** Stabilize layer tests for Adjust operations on all platforms **Ticket:** TBD --------- Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 40bd0ee - Browse repository at this point
Copy the full SHA 40bd0eeView commit details -
[TF FE] Fix LeakyRelu translator and stabilize tests for it (openvino…
…toolkit#26371) **Details:** Handle correctly scalar case for LeakyRelu. **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 359943d - Browse repository at this point
Copy the full SHA 359943dView commit details -
Update flake8-bugbear requirement from <=24.4.26 to <=24.8.19 in /src…
…/bindings/python (openvinotoolkit#26353) Updates the requirements on [flake8-bugbear](https://github.com/PyCQA/flake8-bugbear) to permit the latest version. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/PyCQA/flake8-bugbear/releases">flake8-bugbear's releases</a>.</em></p> <blockquote> <h2>24.8.19</h2> <ul> <li>B910: implement to suggest using Counter() instead of defaultdict(int) (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/489">#489</a>)</li> <li>B901: Do not trigger with explicit Generator return type (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/481">#481</a>)</li> <li>B008: add some comments, rename b008_extend_immutable_calls (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/476">#476</a>)</li> <li>B040: exception with note added not reraised or used (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/477">#477</a>)</li> <li>B039, Add <code>ContextVar</code> with mutable literal or function call as default</li> <li>B040: Add Exception with added note not reraised. (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/474">#474</a>)</li> <li>Run tests in Python 3.13</li> <li>Type annotated code (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/481">#481</a> + <a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/483">#483</a>)</li> <li>Replace hash with unsafe_hash (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/486">#486</a>)</li> </ul> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/2f4e42fd92c933f114305df4a7f42c23673871b2"><code>2f4e42f</code></a> Update CHANGES + move to version 24.8.19 to release</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/c50e9dc34c3add0cf2fb0499d6e32761c7aae6ea"><code>c50e9dc</code></a> [pre-commit.ci] pre-commit autoupdate (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/487">#487</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/f0eeda871eee13b60071c39bd7f83af397589ed2"><code>f0eeda8</code></a> Replace hash with unsafe_hash (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/486">#486</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/593991086069cb3cd347622770c10453734ef582"><code>5939910</code></a> Implemented <a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/323">#323</a>: Added new check B910 to suggest using Counter() instead of ...</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/33993a466fb260d9bcfc82d085b944d8bc003302"><code>33993a4</code></a> [pre-commit.ci] pre-commit autoupdate (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/484">#484</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/5ccce79096f1e42a2cc391c4b84c47d17caea735"><code>5ccce79</code></a> More types (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/483">#483</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/cfda1e8ac827b70ed8d04490c22b859a7a9d9d29"><code>cfda1e8</code></a> Do not trigger B901 with explicit Generator return type (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/481">#481</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/b15feed439fce2919aa4c685d9bd332f11992b67"><code>b15feed</code></a> Set up mypy (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/482">#482</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/3157b89d4168bc8264be3d2ca0a3c556c8d063fe"><code>3157b89</code></a> add b040: exception with note added not reraised or used (<a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/477">#477</a>)</li> <li><a href="https://github.com/PyCQA/flake8-bugbear/commit/188eab867ab068b11ba838ea51d7884c537a96e0"><code>188eab8</code></a> Merge pull request <a href="https://redirect.github.com/PyCQA/flake8-bugbear/issues/476">#476</a> from jakkdl/contextvar_mutable_or_call_default</li> <li>Additional commits viewable in <a href="https://github.com/PyCQA/flake8-bugbear/compare/16.4.1...24.8.19">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>
Configuration menu - View commit details
-
Copy full SHA for 24cb03f - Browse repository at this point
Copy the full SHA 24cb03fView commit details -
Updated documentation on onnx quantization (openvinotoolkit#26364)
### Details: - Updated documentation on onnx quantization ### Tickets: - openvinotoolkit#26192 Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0142a76 - Browse repository at this point
Copy the full SHA 0142a76View commit details -
[DOCS] Release version update for 24.4 release - master (openvinotool…
…kit#26358) Port from openvinotoolkit#26334 Update of links and release version in docs for release 24.4.
Configuration menu - View commit details
-
Copy full SHA for 0982bc9 - Browse repository at this point
Copy the full SHA 0982bc9View commit details -
[GPU] Add indices constant check for crop primitive selection in Crea…
…teGatherOpBase (openvinotoolkit#26361) ### Details: - Add indices constant check for crop primitive selection in CreateGatherOpBase ### Tickets: - 150120
Configuration menu - View commit details
-
Copy full SHA for 39c7e24 - Browse repository at this point
Copy the full SHA 39c7e24View commit details -
[NPU] Fix coverity issue (openvinotoolkit#26374)
### Details: Duplicate with openvinotoolkit#26373 on master branch ### Tickets:
Configuration menu - View commit details
-
Copy full SHA for 7330240 - Browse repository at this point
Copy the full SHA 7330240View commit details -
checkout omz (openvinotoolkit#26382)
### Details: - PR to release openvinotoolkit#26337 - fixes for pip-conflict py3.12 and trivy scan ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 7a130ee - Browse repository at this point
Copy the full SHA 7a130eeView commit details -
[Snippets] Supported dynamic case in SoftmaxReshapeElimination (openv…
…inotoolkit#26350) ### Details: - *Added dynamic case support in the pass "SoftmaxReshapeElimination" when last (reduction) dimension is static - aligned with "MHATokenization" pass* ![image](https://github.com/user-attachments/assets/f160e615-321f-4480-a436-f53055b0ca47) - *Covered by unet test* ### Tickets: - *151134*
Configuration menu - View commit details
-
Copy full SHA for 00cd04d - Browse repository at this point
Copy the full SHA 00cd04dView commit details -
[PT FE] Run EDSR tests in parallel (openvinotoolkit#26395)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 1dc4a18 - Browse repository at this point
Copy the full SHA 1dc4a18View commit details -
[PT FE] Update LLm hub tests (openvinotoolkit#26366)
### Details: - *Update LLm hub tests* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for a4a0f87 - Browse repository at this point
Copy the full SHA a4a0f87View commit details -
[PT FE] Update torchvision tests with quantized and detection models (o…
…penvinotoolkit#26397) ### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for d4ade56 - Browse repository at this point
Copy the full SHA d4ade56View commit details -
[JAX FE] Update JAX version to latest one for testing (openvinotoolki…
…t#26100) **Details:** It is required for validation of Python 3.12 support. **Tickets:** 150860 --------- Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8274132 - Browse repository at this point
Copy the full SHA 8274132View commit details
Commits on Sep 4, 2024
-
[PyOV]: Support direct comparison of PartialShape and Shape with list…
… and tuple (openvinotoolkit#26261) ### Details: - Support direct comparison of PartialShape and Shape with list and tuple - Tests I'm not entirely sure of the dynamic case and could use a hint. ;) Thanks! ### Tickets: - 128813 --------- Co-authored-by: Anastasia Kuporosova <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for cacb820 - Browse repository at this point
Copy the full SHA cacb820View commit details -
[GPU] out-of-bound access of convolution kernel (openvinotoolkit#26326)
### Details: ### Tickets: - 150360
Configuration menu - View commit details
-
Copy full SHA for 4655dd0 - Browse repository at this point
Copy the full SHA 4655dd0View commit details -
[NPU] Remove dKmb support from the plugin (openvinotoolkit#26079)
### Details: - *Remove dKmb support from plugin* ### Tickets: - *CVS-150846*
Configuration menu - View commit details
-
Copy full SHA for d9d1903 - Browse repository at this point
Copy the full SHA d9d1903View commit details -
GHA: removed hardcoded build type (openvinotoolkit#26401)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 9c41f10 - Browse repository at this point
Copy the full SHA 9c41f10View commit details -
[Snippets][CPU] Set N_blk=24 for BRGEMM on AVX2 (openvinotoolkit#26319)
### Details: - *Set `n_blk=24` on avx2 for `BrgemmCPU`* ### Tickets: - *151064*
Configuration menu - View commit details
-
Copy full SHA for a0fe89a - Browse repository at this point
Copy the full SHA a0fe89aView commit details -
[GPU] Improve paddings representation (openvinotoolkit#25716)
### Details: - *rework padding to make get_linear_size() easier thus improve memory pool efficiency and pave path for tensor clear up* - *...* ### Tickets: - *146712*
Configuration menu - View commit details
-
Copy full SHA for 8d44733 - Browse repository at this point
Copy the full SHA 8d44733View commit details -
[TRANSFORMATIONS] Fix Alibi slopes handling for Jais-13b model (openv…
…inotoolkit#26338) [TRANSFORMATIONS] Fix Alibi slopes handling for Jais-13b model The Jais-13b model uses different approach for forming the attention mask using negative values of Alibi slopes for positional embedding. The current implementation of extracting the Alibi slopes and inserting it to PA results into model producing incorrect output. Detect the Jais-13b-like pattern with the negative Alibi constant and multiply it by -1 to use the "raw" (positive) Alibi values before inserting into PA. ### Tickets: - CVS-142320 Signed-off-by: Andrii Staikov <[email protected]> --------- Signed-off-by: Andrii Staikov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5c11661 - Browse repository at this point
Copy the full SHA 5c11661View commit details -
[TF FE] Extend JAX layer tests precommit via TF FE path (openvinotool…
…kit#26409) **Details:** We ensure that more JAX operations are supported via TF FE path **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for aeb034b - Browse repository at this point
Copy the full SHA aeb034bView commit details -
[OVJS] Avoid of usage global variables in tests (openvinotoolkit#26393)
### Details: - Avoid dependencies between tests and unexpected behaviours - Add `beforeEach` ### Tickets: - [CVS-146340](https://jira.devtools.intel.com/browse/CVS-146340) --------- Co-authored-by: Vishniakov Nikolai <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 41394bb - Browse repository at this point
Copy the full SHA 41394bbView commit details -
[GPU] Fix ocl build confliction of MAX() in scatter elements update r…
…ef kernel (openvinotoolkit#26406) ### Details: - Fix ocl build confliction of MAX() in scatter elements update ref kernel. It causes ocl build error when other kernel which uses MAX() macro is generated together in same batch. Scatter elements update ref kernel used #define MAX again. ### Tickets: - 151113
Configuration menu - View commit details
-
Copy full SHA for 6f3f4cc - Browse repository at this point
Copy the full SHA 6f3f4ccView commit details -
[GPU] Fix accuracy issue of Qwen2 model (openvinotoolkit#26411)
### Details: - This PR is a copy of openvinotoolkit#26389 from 2024.4 release branch - It fixes functional issue of Qwen1.5-7b due to unexpected dynamic dimension of query input, and fixes the accuracy of Qwen2 model caused by the order of QK*Scale calculation ### Tickets: - [CVS-151304](https://jira.devtools.intel.com/browse/CVS-151304)
Configuration menu - View commit details
-
Copy full SHA for aed269d - Browse repository at this point
Copy the full SHA aed269dView commit details -
[WHEEL] Split wheel build and build cleanup (openvinotoolkit#26423)
### Details: - Split wheel build and build cleanup - Do not fail build if cleanup step unsuccessful ### Tickets: - *150518*
Configuration menu - View commit details
-
Copy full SHA for 4219701 - Browse repository at this point
Copy the full SHA 4219701View commit details -
[GHA] Add openvino-provider action (openvinotoolkit#26050)
### Details: Usage example https://github.com/openvinotoolkit/private_testrepo/actions/runs/10473510381/job/29005526267?pr=75 ### Tickets: - 147785
Configuration menu - View commit details
-
Copy full SHA for f4c3f88 - Browse repository at this point
Copy the full SHA f4c3f88View commit details -
[GPU] Disable transpose fusion in case of no transpose found on both …
…branches (openvinotoolkit#26429) ### Details: - Avoid unwanted MatMul->Gemm conversion for static shapes as MatMul handler can produce more performance subgraph - Fixes perf regression for stable diffusion unet on non-xmx platforms ### Tickets: - *CVS-151481*
Configuration menu - View commit details
-
Copy full SHA for 266ce7b - Browse repository at this point
Copy the full SHA 266ce7bView commit details -
[conformance] Mark DEVICE_ID mandatory for HW plugin (openvinotoolkit…
…#26297) ### Details: - *DEVICE_ID is optional for SW plugin and mandatory for HW plugin* ### Tickets: - *N/A* --------- Co-authored-by: Chen Peter <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for f5513c6 - Browse repository at this point
Copy the full SHA f5513c6View commit details -
Fixed macOS GHA build (openvinotoolkit#26431)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 88a0ee0 - Browse repository at this point
Copy the full SHA 88a0ee0View commit details
Commits on Sep 5, 2024
-
[GPU] minor fix for paddings (openvinotoolkit#26433)
### Details: - *legacy of openvinotoolkit#25716 - *...* ### Tickets: - *146712*
Configuration menu - View commit details
-
Copy full SHA for 9c71383 - Browse repository at this point
Copy the full SHA 9c71383View commit details -
[GPU] fix for duplicated connection of fusing (openvinotoolkit#26381)
### Details: - A node may be fused into an eltwise node where input0 and input1 connections are the same. - In such case, both input should be regarded ORIGINAL connection. There is no external dependency. ### Tickets: - 26259 (github issue)
Configuration menu - View commit details
-
Copy full SHA for 63fb0f4 - Browse repository at this point
Copy the full SHA 63fb0f4View commit details -
[CPU] fix QKVProjection fake-matching for accuracy with bert models (o…
…penvinotoolkit#26407) ### Details: - prevent QKVProjFusion from mistakenly fusing patterns with integer input (happens in bert-INT8 models with smooth-quant) ### Tickets: - *CVS-151213*
Configuration menu - View commit details
-
Copy full SHA for fd9eab6 - Browse repository at this point
Copy the full SHA fd9eab6View commit details -
[CPU] Add interface to release compiled model internal memory (openvi…
…notoolkit#26262) ### Details: Port openvinotoolkit#26390 to master ### Tickets: - CVS-145873
Configuration menu - View commit details
-
Copy full SHA for ff5a463 - Browse repository at this point
Copy the full SHA ff5a463View commit details -
[TF FE] Stabilize layer tests for ComplexFFT ops and ComplexAbs on al…
…l platforms (openvinotoolkit#26425) **Details:** Stabilize layer tests for ComplexFFT ops and ComplexAbs on all platforms **Tickets:** TBD --------- Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 7cc357c - Browse repository at this point
Copy the full SHA 7cc357cView commit details -
[ONNX] Disabled nonmaxsuppression test for Template plugin due to r2r…
… issues (openvinotoolkit#26386) ### Details: - Disabled test for Template plugin which fails due to r2r issues ### Tickets: - 122120
Configuration menu - View commit details
-
Copy full SHA for 4127181 - Browse repository at this point
Copy the full SHA 4127181View commit details -
[intel-npu] Additional documentation for NPU_MAX_TILES property's usa… (
openvinotoolkit#26435) port: openvinotoolkit#26385 Co-authored-by: Attila Csok <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 485431f - Browse repository at this point
Copy the full SHA 485431fView commit details -
[DOCS] New NNCF article for user documentation - master (openvinotool…
…kit#25981) Addition of new NNCF article for user documentation. --------- Co-authored-by: Karol Blaszczak <[email protected]> Co-authored-by: Alexander Kozlov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5ba9884 - Browse repository at this point
Copy the full SHA 5ba9884View commit details -
[LPT] Handle scale with convert in MatMulTransformation (openvinotool…
…kit#26398) ### Details: This PR handles Matmul dequantization subgraphs with convert on scales. In general, LPT don't expect Convert on dequantization scale: the convert is usually folded at the previous transformation pipeline steps. However, such subgraphs may arise on Matmul weights (in CPU plugin) since decompression handling logic keeps these subgraphs unchanged. Since decompression handling logic concerns only MatMuls, the changes in MatMul LPT is enough. ### Tickets: - *CVS-151589*
Configuration menu - View commit details
-
Copy full SHA for 20ee134 - Browse repository at this point
Copy the full SHA 20ee134View commit details -
[LPT] separateInStandaloneBranch fix (openvinotoolkit#26333)
### Details: - *Added `normalizeDequantization` in `separateInStandaloneBranch`: it is needed since `separateInStandaloneBranch` works correctly only with normalized dequantization* - *Handle all inputs (not only 0's as before) in `separateInStandaloneBranch`* - *Removed `normalizeDequantization` in transformations since this logic is placed inside `separateInStandaloneBranch` now* ### Tickets: - *CVS-150001*
Configuration menu - View commit details
-
Copy full SHA for 8a604a6 - Browse repository at this point
Copy the full SHA 8a604a6View commit details -
[DOCS] Update NNCF weight compression LLM example (openvinotoolkit#25287
) - Update example to not compile the full precision model and not compress the model to INT8 before compressing to INT4 - Update installation command: nncf is now included in optimum[openvino] We may also want to update the optimum example to use OVWeightQuantizationConfig. @AlexKoff88 Co-authored-by: Alexander Kozlov <[email protected]> Co-authored-by: Karol Blaszczak <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e7ae737 - Browse repository at this point
Copy the full SHA e7ae737View commit details -
[GPU] Fix lack of support for 3D (and smaller) shapes in TransposeFus…
…ion pass (openvinotoolkit#26441) ### Details: - Fix lack of support for 3D (and smaller) shapes in TransposeFusion pass - Update IncreasePositionIdsPrecision to support both MatMul and Gemm operations - Copy of openvinotoolkit#26440 ### Tickets: - [CVS-146889](https://jira.devtools.intel.com/browse/CVS-146889)
Configuration menu - View commit details
-
Copy full SHA for aeb942e - Browse repository at this point
Copy the full SHA aeb942eView commit details -
Properly use OneCore toolchain in Windows CC (openvinotoolkit#26446)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 8f84c0d - Browse repository at this point
Copy the full SHA 8f84c0dView commit details -
Fixed extensions (tokenizers) loading on Android (openvinotoolkit#26436)
### Details: - Ported openvinotoolkit#26437
Configuration menu - View commit details
-
Copy full SHA for e2cfe28 - Browse repository at this point
Copy the full SHA e2cfe28View commit details -
[TF FE] Change a root-cause for failing cases (openvinotoolkit#26443)
**Details:** Established a bug in TF itself. **Ticket:** 124436 Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 9969f9f - Browse repository at this point
Copy the full SHA 9969f9fView commit details -
Remove changes to CMAKE_lang_FLAGS_build_type about MT/MTd (openvinot…
…oolkit#26434) ### Details: - *item1* - *...* ### Tickets: - *ticket-id* Signed-off-by: Kang Wenjing <[email protected]> Co-authored-by: Ilya Lavrenov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for ab75969 - Browse repository at this point
Copy the full SHA ab75969View commit details -
[GPU] Add reqd_work_group_size attribute to Groupnorm BFYX OPT kernel (…
…openvinotoolkit#26438) ### Details: - Performance improvement ### Tickets: - 151610
Configuration menu - View commit details
-
Copy full SHA for 820bc89 - Browse repository at this point
Copy the full SHA 820bc89View commit details -
[GHA] Default registry to dockerhub (openvinotoolkit#26448)
Makes local image building easier ### Tickets: - 149907
Configuration menu - View commit details
-
Copy full SHA for 22079ca - Browse repository at this point
Copy the full SHA 22079caView commit details
Commits on Sep 6, 2024
-
[PyOV] remove deprecated api (openvinotoolkit#26442)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 1e1787c - Browse repository at this point
Copy the full SHA 1e1787cView commit details -
[PyOV] Test export/import (openvinotoolkit#26311)
### Details: - *item1* - *...* ### Tickets: - *[CVS-133903](https://jira.devtools.intel.com/browse/CVS-133903)*
Configuration menu - View commit details
-
Copy full SHA for 5807ed4 - Browse repository at this point
Copy the full SHA 5807ed4View commit details -
[DOCS] Updating NodeJS API docs (openvinotoolkit#26445)
Adding info on: `Model.getOutputElementType`, `Model.clone`, `CompiledModel.getProperty`, `CompiledModel.setProperty`.
Configuration menu - View commit details
-
Copy full SHA for d481f34 - Browse repository at this point
Copy the full SHA d481f34View commit details -
Revise Transformation docs: python API (openvinotoolkit#26008)
### Details: - MatcherPass doc - ModelPass doc ### Tickets: - CVS-76333 --------- Co-authored-by: Maciej Smyk <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 86dfbca - Browse repository at this point
Copy the full SHA 86dfbcaView commit details -
[GPU] Support bfyx and fsv32 input formats for concat (openvinotoolki…
…t#26372) ### Details: - *Implement shape agnostic concat ref kernel* - *The shape agnostic concat ref kernel supports bfyx and fsv32 input formats* ### Tickets: - *149462*
Configuration menu - View commit details
-
Copy full SHA for fe87a96 - Browse repository at this point
Copy the full SHA fe87a96View commit details -
[CPU] [Snippets] Implement Convert for Snippets on ARM (openvinotoolk…
…it#25815) ### Details: - *Add jit implementation for Convert emitters on ARM* - *Add jit implementation for Load/Store emitters for precision i32, f16, i8, u8 on ARM* - *Add Snippets tokenization for Convert on ARM* - *Enable LoadConvertSaturation and three other counterparts* - *Test case coverage* ### Tickets: - *[CVS-141288](https://jira.devtools.intel.com/browse/CVS-141288)* - *[CVS-141294](https://jira.devtools.intel.com/browse/CVS-141294)*
Configuration menu - View commit details
-
Copy full SHA for 48a6777 - Browse repository at this point
Copy the full SHA 48a6777View commit details -
Configuration menu - View commit details
-
Copy full SHA for 65638ce - Browse repository at this point
Copy the full SHA 65638ceView commit details -
[PT FE] Use static seed for quantized tests (openvinotoolkit#26451)
### Details: - *Use static seed for quantized tests* ### Tickets: - *CVS-151071*
Configuration menu - View commit details
-
Copy full SHA for 5f0fc54 - Browse repository at this point
Copy the full SHA 5f0fc54View commit details -
[intel-npu] Add functional tests to check correct memory deallocation (…
…openvinotoolkit#25980) ### Details: - *The following new test cases are being added:* ``` 1) allocate / de-allocate a model for N times 2) allocate a model create infer request / destroy infer request for N times de-allocate and destroy the model 3) allocate a model create infer request set_tensort(with new buffer every time) for N times destroy infer request de-allocate and destroy the model 4) allocate a model create infer request local_tensor = infer_request.get_output_tensor() destroy infer request // output tensor should not be deallocated here de-allocate and destroy the model destroy local_tensor // output tensor should be deallocated here ``` ### Tickets: - *E129376*
Configuration menu - View commit details
-
Copy full SHA for 69180e2 - Browse repository at this point
Copy the full SHA 69180e2View commit details -
[LPT] Handle Gather in markup transformation (openvinotoolkit#26428)
### Details: When GatherTransformation was implemented, the necessary handling in LPT Markup transformations was skipped. This PR adds gather layers handling in LPT Markup transformations ### Tickets: - *CVS-151590*
Configuration menu - View commit details
-
Copy full SHA for e3bc678 - Browse repository at this point
Copy the full SHA e3bc678View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1b98e2d - Browse repository at this point
Copy the full SHA 1b98e2dView commit details -
Update quantization-aware-training-pytorch.rst (openvinotoolkit#26414)
The suggested learning rate in this document seems wrong. It should be 1e-5 rather than 10e-5. ### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 626571d - Browse repository at this point
Copy the full SHA 626571dView commit details -
[Snippets] Enable libxsmm GHA job (openvinotoolkit#26356)
### Details: - *Enable functional and unit tests in the Libxsmm GHA job* ### Tickets: - *151223*
Configuration menu - View commit details
-
Copy full SHA for c5892bd - Browse repository at this point
Copy the full SHA c5892bdView commit details -
add failed model to the skip list (openvinotoolkit#26365)
### Details: - added cientnet/b2/feature-vector to skip list due to url error ### Tickets: - 151259
Configuration menu - View commit details
-
Copy full SHA for 102e875 - Browse repository at this point
Copy the full SHA 102e875View commit details
Commits on Sep 9, 2024
-
[DOC] add cache_encryption_callbacks example in guide (openvinotoolki…
…t#26432) ### Details: - *add example of cache_encryption_callbacks in document* ### Tickets: - *CVS-123336* --------- Co-authored-by: Chen Peter <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 91b5744 - Browse repository at this point
Copy the full SHA 91b5744View commit details -
[TF FE] Fix translator for LinSpace operation and stabilize tests for…
… it on all platforms (openvinotoolkit#26477) **Details:** Fix translator for LinSpace operation and stabilize tests for it on all platforms **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d25367c - Browse repository at this point
Copy the full SHA d25367cView commit details -
[TF FE] Stabilize layer tests for Xlogy and Xlog1py on all platforms (o…
…penvinotoolkit#26474) **Details:** Stabilize layer tests for Xlogy and Xlog1py on all platforms **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 556a28c - Browse repository at this point
Copy the full SHA 556a28cView commit details -
[GPU] Update gpu passes to keep valid layouts more often (openvinotoo…
…lkit#26426) ### Details: - Added explicit layout recalc after some graph modifications to avoid full invalidation of layouts later on random attempt to call `get_output_layout()` - Remove i64 data type handling in add_required_reorder pass as ConvertPrecision is supposed to convert everything into i32. - Move shape info tensor alloc to `update_shape_info_tensor` call to avoid allocation in case if the buffer is not needed for selected impl - Fixed impl forcing in some unit tests + other minor test changes
Configuration menu - View commit details
-
Copy full SHA for 94a9675 - Browse repository at this point
Copy the full SHA 94a9675View commit details -
[TF FE]: Support complex tensors for Prod operations (openvinotoolkit…
…#26475) ### Details: - Fixed fifth case of `atan2` implementation (it returned `pi/2` instead of `-pi/2`). - Moved `atan2` to `utils`. - Create helper functions for converting complex number from rectangular to polar form and the other way around. - Support complex tensors for `Prod` operations + unit tests. ### Tickets: - [None](openvinotoolkit#23233) ### Resources used: - https://math.stackexchange.com/questions/1938894/imaginary-part-of-a-product-of-n-complex-numbers - https://en.m.wikipedia.org/wiki/Euler%27s_formula --------- Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for a8b4e91 - Browse repository at this point
Copy the full SHA a8b4e91View commit details -
[GPU] Fix random tests (openvinotoolkit#26486)
* Fix out of bound memory in tests cases, phase1
Configuration menu - View commit details
-
Copy full SHA for cea6168 - Browse repository at this point
Copy the full SHA cea6168View commit details -
Add USE_TBBBIND_2_5 before the definition of binding_observer (openvi…
…notoolkit#26482) ### Details: - *`binding_observer` is defined but not implemented when USE_TBBBIND_2_5 is not defined. So add `USE_TBBBIND_2_5` before the definition of `binding_observer`* ### Tickets: - *151706*
Configuration menu - View commit details
-
Copy full SHA for 420c97f - Browse repository at this point
Copy the full SHA 420c97fView commit details -
[PT FE] Update model hub tests (openvinotoolkit#26467)
### Details: - *Update model hub tests* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for aa676ca - Browse repository at this point
Copy the full SHA aa676caView commit details -
[PT FE] Update torch to 2.4.1 in tests (openvinotoolkit#26464)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for a5f7a34 - Browse repository at this point
Copy the full SHA a5f7a34View commit details -
[GPU] fix zero array in network unit test (openvinotoolkit#26470)
### Details: - Check for zero intermediate buffers ### Tickets: - N/A
Configuration menu - View commit details
-
Copy full SHA for a4ac961 - Browse repository at this point
Copy the full SHA a4ac961View commit details -
[GHA] Enable more convenient OV provider call from other repos (openv…
…inotoolkit#26471) To call the action via a single step "- uses: openvinotoolkit/openvino/.github/actions/openvino_provider@master" w/o extra checkout
Configuration menu - View commit details
-
Copy full SHA for 4a5b2f4 - Browse repository at this point
Copy the full SHA 4a5b2f4View commit details -
[DOCS] GenAI Docker images note (openvinotoolkit#26489)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for f2bea83 - Browse repository at this point
Copy the full SHA f2bea83View commit details -
Docs change csv to json file in benchmark graphs (openvinotoolkit#26494)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 7557de4 - Browse repository at this point
Copy the full SHA 7557de4View commit details -
Configuration menu - View commit details
-
Copy full SHA for 796b75e - Browse repository at this point
Copy the full SHA 796b75eView commit details -
[DOCS] New Key Features in docs - master (openvinotoolkit#26308)
New Key Features article for the documentation. --------- Co-authored-by: Karol Blaszczak <[email protected]> Co-authored-by: Sebastian Golebiewski <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 996f52b - Browse repository at this point
Copy the full SHA 996f52bView commit details -
[DOCS] Removing redundant data from articles. (openvinotoolkit#26496)
Removing old ID labels from articles.
Configuration menu - View commit details
-
Copy full SHA for 00e5635 - Browse repository at this point
Copy the full SHA 00e5635View commit details -
[Snippets][CPU] Disabled dynamic MHA tokenization if rtCache is not u…
…sed (openvinotoolkit#26376) ### Details: - *To reduce overheads of ShapeInference and CodeGeneration of dynamic Subgraphs, CPU node Subgraph uses Runtime Cache of the plugin. If Runtime Cache capacity is zero, dynamic subgraphs should not be tokenized - it will lead to performance degradations. This PR disables dynamic MHA tokenization if `config.rtCacheCapacity == 0`* ### Tickets: - *150951*
Configuration menu - View commit details
-
Copy full SHA for a87851d - Browse repository at this point
Copy the full SHA a87851dView commit details
Commits on Sep 10, 2024
-
[GPU] Added support fsv16/32 depthwise shape agnostic convolution (op…
…envinotoolkit#26402) ### Tickets: - *[148550](https://jira.devtools.intel.com/browse/CVS-148550)*
Configuration menu - View commit details
-
Copy full SHA for 1a6c1b6 - Browse repository at this point
Copy the full SHA 1a6c1b6View commit details -
[Paddle FE] Paddle frontend op interpolate support SizeTensor (openvi…
…notoolkit#26478) ### Details: - *Paddle frontend op interpolate support SizeTensor* ### Tickets: - *CVS-150557*
Configuration menu - View commit details
-
Copy full SHA for 49b4db3 - Browse repository at this point
Copy the full SHA 49b4db3View commit details -
Fix typo in EnvVar class and fix filtering functionality (openvinotoo…
…lkit#26459) ### Details: Now EnvVar works correctly in case of bool values Thanks @ceciliapeng2011 for finding + updated the description and the filtering functionality ### Tickets: - *N/A*
Configuration menu - View commit details
-
Copy full SHA for cfecacd - Browse repository at this point
Copy the full SHA cfecacdView commit details -
[GPU] Simplify items_num and leftovers calculation (openvinotoolkit#2…
…6495) ### Details: - Simplify and reduce code - Very slight performance improvement for dynamic shapes ### Tickets: -
Configuration menu - View commit details
-
Copy full SHA for 105f44d - Browse repository at this point
Copy the full SHA 105f44dView commit details -
Append input index to input names (openvinotoolkit#26453)
### Details: - `ii.get_index` can return same index for different inputs: ``` (gdb) p inputInfo $1 = std::vector of length 2, capacity 2 = {{ m_node = std::shared_ptr<ov::Node> (use count 3, weak count 2) = {get() = 0x555555872860}, m_index = 0}, { m_node = std::shared_ptr<ov::Node> (use count 3, weak count 2) = {get() = 0x555555873fa0}, m_index = 0}} ``` - This may result in the collision of names. - When different inputs are referred by the same name, single image test will populate only one such input. - Append `id` to the name to assure its uniqueness. ### Tickets: - E-137973
Configuration menu - View commit details
-
Copy full SHA for 52c9ae7 - Browse repository at this point
Copy the full SHA 52c9ae7View commit details -
Updated LLM compression related information (openvinotoolkit#26460)
Co-authored-by: Tatiana Savina <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for a72b4ef - Browse repository at this point
Copy the full SHA a72b4efView commit details -
[JAX FE][DOCS] Document Support of JAX/Flax models by OpenVINO (openv…
…inotoolkit#26513) **Details:** Document Support of JAX/Flax models by OpenVINO **Ticket:** 150970 --------- Signed-off-by: Kazantsev, Roman <[email protected]> Co-authored-by: Karol Blaszczak <[email protected]> Co-authored-by: Sebastian Golebiewski <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8a875c3 - Browse repository at this point
Copy the full SHA 8a875c3View commit details -
[GPU] Fix network c-tor to avoid config mismatch (openvinotoolkit#26485)
### Details: - Network class c-tor had version with both program and config parameters, and this config parameter was actually used which caused mismatch between program and network config values as inside the program we call apply_user_properties() method. - This PR removes Config from this c-tor and get it from program class
Configuration menu - View commit details
-
Copy full SHA for 014e13c - Browse repository at this point
Copy the full SHA 014e13cView commit details -
[TRANSFORMATIONS][DEBUG] Include extra debug info to MatcherPass (ope…
…nvinotoolkit#26499) Add OPENVINO_DEBUG for a matching attempt. Catch possible callback exceptions to include a name of the matcher which is currently being applied Co-authored-by: Ivan Tikhonov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 01fd36e - Browse repository at this point
Copy the full SHA 01fd36eView commit details -
[LPT] NormalizeDequantization correction (openvinotoolkit#26490)
### Details: - *Correct NormalizeDequantization behavior in order to avoid eltwise inputs reshuffle in case if both inputs are constant* - *Added converts on constant path support in NormalizeDequantization* ### Tickets: - *CVS-151869* - *CVS-79740*
Configuration menu - View commit details
-
Copy full SHA for 3e2f1eb - Browse repository at this point
Copy the full SHA 3e2f1ebView commit details -
[Snippets] Dynamic SplitDimensionM leftovers (openvinotoolkit#26354)
### Details: PR with leftovers of openvinotoolkit#25733.
Configuration menu - View commit details
-
Copy full SHA for a04346b - Browse repository at this point
Copy the full SHA a04346bView commit details -
[CPU][ARM] Fix output detection in
ConvertReduceMultiAxisBase
trans……formation (openvinotoolkit#26500) ### Details: - The fix of openvinotoolkit/nncf#2943 - The error `Default output not supported` could be raised in `ConvertReduceMultiAxisBase` transformation because output of Reduce input node was not specified. It works well if the node has default output, but for some nodes (like `VariadicSplit`) it leads to the error. - The fix is to explicitly specify needed output. ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 090da7b - Browse repository at this point
Copy the full SHA 090da7bView commit details
Commits on Sep 11, 2024
-
Update pybind11 version (openvinotoolkit#26520)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 20ef40d - Browse repository at this point
Copy the full SHA 20ef40dView commit details -
[CI] [GHA] Introduce Fedora 29 (openvinotoolkit#26336)
### Tickets: - *148720*
Configuration menu - View commit details
-
Copy full SHA for 79a75a6 - Browse repository at this point
Copy the full SHA 79a75a6View commit details -
[GPU] Fix OOB access in reorder (openvinotoolkit#26522)
### Details: - Fix out of bound access in reorder
Configuration menu - View commit details
-
Copy full SHA for f513f29 - Browse repository at this point
Copy the full SHA f513f29View commit details -
[GPU] Fix unit tests ph2 (openvinotoolkit#26524)
### Details: - Fixes in tests
Configuration menu - View commit details
-
Copy full SHA for efdbad6 - Browse repository at this point
Copy the full SHA efdbad6View commit details -
[GPU] Optimize graph transformation for pytorch (openvinotoolkit#26410)
### Details: - Transpose fusion into MatMul may have caused perf drop even if tensor is aligned by 16 - For small tensors & aligned-by-16, fuse Transpose into MatMul. - For large tensors, do not fuse Transpose - Remove Pad in front of MaxPool. - MaxPool adds padding for CEIL_PYTORCH rounding type. - The pad should be removed if the pad_begin and pad_end are 0. Otherwise, it would cause perf drop. ### Tickets: - *150556*
Configuration menu - View commit details
-
Copy full SHA for f026020 - Browse repository at this point
Copy the full SHA f026020View commit details -
Fix path to generated ov_plugins.hpp (openvinotoolkit#26521)
OpenVino fails to compile if it's used as a subdirectory in CMakeLists.txt. For example, below scenario: ``` $ git clone https://github.com/openvinotoolkit/openvino --recursive $ cat CMakeLists.txt project(testing) add_subdirectory(openvino) $ mkdir build $ cmake -B build -S . $ make -C build openvino_runtime_obj ``` ends with compilation error: ``` core_impl.cpp:32:10: fatal error: ov_plugins.hpp: No such file or directory 32 | #include "ov_plugins.hpp" | ^~~~~~~~~~~~~~~~ ``` because of incorrect include path. openvino_runtime_obj target expects ov_plugins.hpp to be present in ${CMAKE_CURRENT_BINARY_DIR} (which is build/openvino/src/inference in this case), but ov_plugins.hpp is generated to ${CMAKE_BINARY_DIR}/src/inference (which is build/src/inference). To fix that, ov_generate_plugins_hpp can use BINARY_DIR property of openvino_runtime_obj instead of ${CMAKE_BINARY_DIR}/src/inference.
Configuration menu - View commit details
-
Copy full SHA for c5adb2e - Browse repository at this point
Copy the full SHA c5adb2eView commit details -
[GHA] Add ov_wheel_source output for ov provider (openvinotoolkit#26517)
For some cases (e.g. during tokenizers build via pip wheel) we need only wheel source argument, not the input for pip install
Configuration menu - View commit details
-
Copy full SHA for be1120d - Browse repository at this point
Copy the full SHA be1120dView commit details -
[DOCS] Updating references to OMZ in docs (openvinotoolkit#26523)
Updating reference to Open Model Zoo repository.
Configuration menu - View commit details
-
Copy full SHA for 2ad4199 - Browse repository at this point
Copy the full SHA 2ad4199View commit details -
[TF FE] Stabilize tf.keras.layers.Dot layer tests on all platforms (o…
…penvinotoolkit#26528) **Details:** Stabilize tf.keras.layers.Dot layer tests on all platforms **Ticket:** 151986 Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 7192ce1 - Browse repository at this point
Copy the full SHA 7192ce1View commit details -
[Transformations] MatMulMultiplyFusion: added a check on consumers co…
…unt (openvinotoolkit#26525) ### Details: - *Current MatMulMultiplyFusion implementation transforms matmuls with several consumers which is incorrect. This PR adds the corresponding check in transformation's matcher in order to avoid transformation in case of several matmul consumers* ### Tickets: - *N\A* --------- Co-authored-by: Ivan Tikhonov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 5f7056b - Browse repository at this point
Copy the full SHA 5f7056bView commit details -
[DOCS] NNCF section adjustments pass2 (openvinotoolkit#26461)
continuation of: openvinotoolkit#25981
Configuration menu - View commit details
-
Copy full SHA for adffbfd - Browse repository at this point
Copy the full SHA adffbfdView commit details -
[GPU] Optimize RMS Stack Size for Better Performance (openvinotoolkit…
…#26515) ### Details: - Since stack_size can be calculated, we can use smaller stack size instead of 33 to achieve better performance. ### Performance: - qwen2-0.5B’s throughput increased from 40.2 qps to 41.8 qps. ### Tickets: - CVS-152013
Configuration menu - View commit details
-
Copy full SHA for 90d1219 - Browse repository at this point
Copy the full SHA 90d1219View commit details -
[GHA] Artifact folder names consistency check (openvinotoolkit#26538)
### Details: To avoid sudden breaking of CI's in repos retrieving these artifacts via OV provider or Jenkins ### Tickets: - 152261
Configuration menu - View commit details
-
Copy full SHA for 05bd0c7 - Browse repository at this point
Copy the full SHA 05bd0c7View commit details -
[CPU] Introduce SubModel op and Composite node (openvinotoolkit#25385)
with the idea of wrapping up (grouping) parts of the model / graph into a inner model / graph without any (almost) runtime overhead. SubModel op and Compoisite node are expected to perform no extra logic and to only execute an inner model / graph ### Details: - The naming "Composite node and SubModel operation" are discussable. A convenient option for the Composite node would be "SubGraph", but it is taken by snippets, which seems logical as well. - the Composite node is not yet used by plugin - single layer tests are to be added later. The problem is that currently it is impossible to use cpu plugin specific transformations and operations in scope of cpu func tests. - ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 5cba1d5 - Browse repository at this point
Copy the full SHA 5cba1d5View commit details
Commits on Sep 12, 2024
-
[CPU]Mark Range as FP32 (openvinotoolkit#26456)
### Details: - *Make `MarkFloatingPointRange` as public transformation API* - *Use `MarkFloatingPointRange` in CPU BF16 inference* ### Tickets: - *CVS-144062*
Configuration menu - View commit details
-
Copy full SHA for e8f1167 - Browse repository at this point
Copy the full SHA e8f1167View commit details -
[GPU] Fix fc_tiled kernel for eltwise fusion and allow fusion for dyn…
…amic shape (openvinotoolkit#26502) ### Details: - Fixed bug in fc_tiled_opt kernel + eltwise fusion - Allow fc (fp precision) + eltwise fusion for dynamic case & unfuse if unsupported kernel is selected ### Tickets: - CVS-151537
Configuration menu - View commit details
-
Copy full SHA for cbe2edf - Browse repository at this point
Copy the full SHA cbe2edfView commit details -
[GPU] OpenCL headers compatibility comment (openvinotoolkit#26534)
### Details: - *item1* - *...* ### Tickets: - *[CVS-138065](https://jira.devtools.intel.com/browse/CVS-138065)* --------- Co-authored-by: Tatiana Savina <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e611ede - Browse repository at this point
Copy the full SHA e611edeView commit details -
Bump actions/upload-artifact from 4.3.4 to 4.4.0 (openvinotoolkit#26544)
Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4.3.4 to 4.4.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/actions/upload-artifact/releases">actions/upload-artifact's releases</a>.</em></p> <blockquote> <h2>v4.4.0</h2> <h2>Notice: Breaking Changes
⚠️ </h2> <p>We will no longer include hidden files and folders by default in the <code>upload-artifact</code> action of this version. This reduces the risk that credentials are accidentally uploaded into artifacts. Customers who need to continue to upload these files can use a new option, <code>include-hidden-files</code>, to continue to do so.</p> <p>See <a href="https://github.blog/changelog/2024-08-19-notice-of-upcoming-deprecations-and-breaking-changes-in-github-actions-runners/">"Notice of upcoming deprecations and breaking changes in GitHub Actions runners"</a> changelog and <a href="https://redirect.github.com/actions/upload-artifact/issues/602">this issue</a> for more details.</p> <h2>What's Changed</h2> <ul> <li>Exclude hidden files by default by <a href="https://github.com/joshmgross"><code>@joshmgross</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/598">actions/upload-artifact#598</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4.3.6...v4.4.0">https://github.com/actions/upload-artifact/compare/v4.3.6...v4.4.0</a></p> <h2>v4.3.6</h2> <h2>What's Changed</h2> <ul> <li>Revert to <code>@actions/artifact</code> 2.1.8 by <a href="https://github.com/robherley"><code>@robherley</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/594">actions/upload-artifact#594</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4...v4.3.6">https://github.com/actions/upload-artifact/compare/v4...v4.3.6</a></p> <h2>v4.3.5</h2> <h2>What's Changed</h2> <ul> <li>Bump <code>@actions/artifact</code> to v2.1.9 by <a href="https://github.com/robherley"><code>@robherley</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/588">actions/upload-artifact#588</a> <ul> <li>Fixed artifact upload chunk timeout logic <a href="https://redirect.github.com/actions/toolkit/pull/1774">#1774</a></li> <li>Use lazy stream to prevent issues with open file limits <a href="https://redirect.github.com/actions/toolkit/pull/1771">#1771</a></li> </ul> </li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v4.3.4...v4.3.5">https://github.com/actions/upload-artifact/compare/v4.3.4...v4.3.5</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/actions/upload-artifact/commit/50769540e7f4bd5e21e526ee35c689e35e0d6874"><code>5076954</code></a> Merge pull request <a href="https://redirect.github.com/actions/upload-artifact/issues/598">#598</a> from actions/joshmgross/exclude-hidden-files</li> <li><a href="https://github.com/actions/upload-artifact/commit/d52396ac5d312b1bda1b9fe2ae55d862c65abd68"><code>d52396a</code></a> Add a warning about enabling <code>include-hidden-files</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/710f36207581d49e465ccbed3e007c317290fe11"><code>710f362</code></a> Remove "merged" from <code>include-hidden-files</code> input description</li> <li><a href="https://github.com/actions/upload-artifact/commit/3b315f26f6f828f74089e1863e67e60e48e37563"><code>3b315f2</code></a> <code>npm run release</code> again 🙂</li> <li><a href="https://github.com/actions/upload-artifact/commit/3be2180eb79b56341b1b127ca06a5adba2a5a3e4"><code>3be2180</code></a> Remove another trailing comma</li> <li><a href="https://github.com/actions/upload-artifact/commit/453e8d0a40444939146df5febed0b1221b9dd073"><code>453e8d0</code></a> Update glob license</li> <li><a href="https://github.com/actions/upload-artifact/commit/0a398c14804ead69f18a166f0a99f91239261ee5"><code>0a398c1</code></a> <code>npm run release</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/a0c40cf60283f329afac2bc0fa77f1a59fa11e6e"><code>a0c40cf</code></a> Update to latest <code>@actions/glob</code> and fix tests</li> <li><a href="https://github.com/actions/upload-artifact/commit/acb59e47763b6e601a344d04cf3c082ee336b709"><code>acb59e4</code></a> <code>lint</code></li> <li><a href="https://github.com/actions/upload-artifact/commit/cb6558bb10fe4afe4054d0be4b3136e673eb5e7f"><code>cb6558b</code></a> Exclude hidden files by default</li> <li>Additional commits viewable in <a href="https://github.com/actions/upload-artifact/compare/v4.3.4...50769540e7f4bd5e21e526ee35c689e35e0d6874">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/upload-artifact&package-manager=github_actions&previous-version=4.3.4&new-version=4.4.0)](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>Configuration menu - View commit details
-
Copy full SHA for 7fa728b - Browse repository at this point
Copy the full SHA 7fa728bView commit details -
[DOCS] Add verification of json benchmarks files (openvinotoolkit#26527)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 97030ad - Browse repository at this point
Copy the full SHA 97030adView commit details -
[GPU] Fix a bug of onednn asymmetric convolution (openvinotoolkit#26536)
### Details: - *Fixed incorrect value being set when weights zero-points are scalar.* ### Tickets: - *151898*
Configuration menu - View commit details
-
Copy full SHA for 4785767 - Browse repository at this point
Copy the full SHA 4785767View commit details -
[GHA]: added GenAI integration build using extra modules way (openvin…
…otoolkit#26546) ### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 88597e5 - Browse repository at this point
Copy the full SHA 88597e5View commit details -
[OV JS] Expose core.query_model to Node.js Api (openvinotoolkit#25966)
### Details: Changes as part of this PR include: * Add a CoreWrap::query_model function: Calls the underlying core.query_model function * Update the addon.ts with the queryModel method Resolves: openvinotoolkit#24373 ### Tickets: - openvinotoolkit#24373 Signed-off-by: Nashez Zubair <[email protected]> Co-authored-by: Alicja Miloszewska <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 60121df - Browse repository at this point
Copy the full SHA 60121dfView commit details -
Configuration menu - View commit details
-
Copy full SHA for a664d49 - Browse repository at this point
Copy the full SHA a664d49View commit details -
Add ReduceReshape transformation to NopElimination scope (openvinotoo…
…lkit#26427) ### Details: Added ReduceReshape elimination transformation which eliminates Reshape op in some cases for Reduce-Reshape patterns ### Tickets: - *CVS-150952*
Configuration menu - View commit details
-
Copy full SHA for a4f8e52 - Browse repository at this point
Copy the full SHA a4f8e52View commit details -
Update jax requirement from <=0.4.31 to <=0.4.32 in /tests (openvinot…
…oolkit#26548) Updates the requirements on [jax](https://github.com/google/jax) to permit the latest version. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/google/jax/releases">jax's releases</a>.</em></p> <blockquote> <h2>JAX release v0.4.32</h2> <ul> <li> <p>New Functionality</p> <ul> <li>Added <code>jax.extend.ffi.ffi_call</code> and <code>jax.extend.ffi.ffi_lowering</code> to support the use of the new <code>ffi-tutorial</code> to interface with custom C++ and CUDA code from JAX.</li> </ul> </li> <li> <p>Changes</p> <ul> <li><code>jax_enable_memories</code> flag is set to <code>True</code> by default.</li> <li><code>jax.numpy</code> now supports v2023.12 of the Python Array API Standard. See <code>python-array-api</code> for more information.</li> <li>Computations on the CPU backend may now be dispatched asynchronously in more cases. Previously non-parallel computations were always dispatched synchronously. You can recover the old behavior by setting <code>jax.config.update('jax_cpu_enable_async_dispatch', False)</code>.</li> <li>Added new <code>jax.process_indices</code> function to replace the <code>jax.host_ids()</code> function that was deprecated in JAX v0.2.13.</li> <li>To align with the behavior of <code>numpy.fabs</code>, <code>jax.numpy.fabs</code> has been modified to no longer support <code>complex dtypes</code>.</li> <li><code>jax.tree_util.register_dataclass</code> now checks that <code>data_fields</code> and <code>meta_fields</code> includes all dataclass fields with <code>init=True</code> and only them, if <code>nodetype</code> is a dataclass.</li> <li>Several <code>jax.numpy</code> functions now have full <code>jax.numpy.ufunc</code> interfaces, including <code>jax.numpy.add</code>, <code>jax.numpy.multiply</code>, <code>jax.numpy.bitwise_and</code>, <code>jax.numpy.bitwise_or</code>, <code>jax.numpy.bitwise_xor</code>, <code>jax.numpy.logical_and</code>, <code>jax.numpy.logical_and</code>, and <code>jax.numpy.logical_and</code>. In future releases we plan to expand these to other ufuncs.</li> <li>Added <code>jax.lax.optimization_barrier</code>, which allows users to prevent compiler optimizations such as common-subexpression elimination and to control scheduling.</li> </ul> </li> <li> <p>Breaking changes</p> <ul> <li>The MHLO MLIR dialect (<code>jax.extend.mlir.mhlo</code>) has been removed. Use the <code>stablehlo</code> dialect instead.</li> </ul> </li> <li> <p>Deprecations</p> <ul> <li>Complex inputs to <code>jax.numpy.clip</code> and {func}<code>jax.numpy.hypot</code> are no longer allowed, after being deprecated since JAX v0.4.27.</li> <li>Deprecated the following APIs: <ul> <li><code>jax.lib.xla_bridge.xla_client</code>: use <code>jax.lib.xla_client</code> directly.</li> <li><code>jax.lib.xla_bridge.get_backend</code>: use <code>jax.extend.backend.get_backend</code>.</li> <li><code>jax.lib.xla_bridge.default_backend</code>: use <code>jax.extend.backend.default_backend</code>.</li> </ul> </li> <li>The <code>jax.experimental.array_api</code> module is deprecated, and importing it is no longer required to use the Array API. <code>jax.numpy</code> supports the array API directly; see <code>python-array-api</code> for more information.</li> <li>The internal utilities <code>jax.core.check_eqn</code>, <code>jax.core.check_type</code>, and <code>jax.core.check_valid_jaxtype</code> are now deprecated, and will be removed in the future.</li> <li><code>jax.numpy.round_</code> has been deprecated, following removal of the corresponding</li> </ul> </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/google/jax/blob/main/CHANGELOG.md">jax's changelog</a>.</em></p> <blockquote> <h2>jax 0.4.32 (September 11, 2024)</h2> <ul> <li> <p>New Functionality</p> <ul> <li>Added {func}<code>jax.extend.ffi.ffi_call</code> and {func}<code>jax.extend.ffi.ffi_lowering</code> to support the use of the new {ref}<code>ffi-tutorial</code> to interface with custom C++ and CUDA code from JAX.</li> </ul> </li> <li> <p>Changes</p> <ul> <li><code>jax_pmap_no_rank_reduction</code> flag is set to <code>True</code> by default. <ul> <li>array[0] on a pmap result now introduces a reshape (use array[0:1] instead).</li> <li>The per-shard shape (accessable via jax_array.addressable_shards or jax_array.addressable_data(0)) now has a leading (1, ...). Update code that directly accesses shards accordingly. The rank of the per-shard-shape now matches that of the global shape which is the same behavior as jit. This avoids costly reshapes when passing results from pmap into jit.</li> </ul> </li> <li><code>jax_enable_memories</code> flag is set to <code>True</code> by default.</li> <li>{mod}<code>jax.numpy</code> now supports v2023.12 of the Python Array API Standard. See {ref}<code>python-array-api</code> for more information.</li> <li>Computations on the CPU backend may now be dispatched asynchronously in more cases. Previously non-parallel computations were always dispatched synchronously. You can recover the old behavior by setting <code>jax.config.update('jax_cpu_enable_async_dispatch', False)</code>.</li> <li>Added new {func}<code>jax.process_indices</code> function to replace the <code>jax.host_ids()</code> function that was deprecated in JAX v0.2.13.</li> <li>To align with the behavior of <code>numpy.fabs</code>, <code>jax.numpy.fabs</code> has been modified to no longer support <code>complex dtypes</code>.</li> <li><code>jax.tree_util.register_dataclass</code> now checks that <code>data_fields</code> and <code>meta_fields</code> includes all dataclass fields with <code>init=True</code> and only them, if <code>nodetype</code> is a dataclass.</li> <li>Several {mod}<code>jax.numpy</code> functions now have full {class}<code>~jax.numpy.ufunc</code> interfaces, including {obj}<code>~jax.numpy.add</code>, {obj}<code>~jax.numpy.multiply</code>, {obj}<code>~jax.numpy.bitwise_and</code>, {obj}<code>~jax.numpy.bitwise_or</code>, {obj}<code>~jax.numpy.bitwise_xor</code>, {obj}<code>~jax.numpy.logical_and</code>, {obj}<code>~jax.numpy.logical_and</code>, and {obj}<code>~jax.numpy.logical_and</code>. In future releases we plan to expand these to other ufuncs.</li> <li>Added {func}<code>jax.lax.optimization_barrier</code>, which allows users to prevent compiler optimizations such as common-subexpression elimination and to control scheduling.</li> </ul> </li> <li> <p>Breaking changes</p> <ul> <li>The MHLO MLIR dialect (<code>jax.extend.mlir.mhlo</code>) has been removed. Use the <code>stablehlo</code> dialect instead.</li> </ul> </li> <li> <p>Deprecations</p> <ul> <li>Complex inputs to {func}<code>jax.numpy.clip</code> and {func}<code>jax.numpy.hypot</code> are no longer allowed, after being deprecated since JAX v0.4.27.</li> <li>Deprecated the following APIs: <ul> <li><code>jax.lib.xla_bridge.xla_client</code>: use {mod}<code>jax.lib.xla_client</code> directly.</li> <li><code>jax.lib.xla_bridge.get_backend</code>: use {func}<code>jax.extend.backend.get_backend</code>.</li> </ul> </li> </ul> </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/google/jax/commit/1594d2f30fdbfebf693aba4a2b264e4a3e52acc6"><code>1594d2f</code></a> Prepare for v0.4.32 release.</li> <li><a href="https://github.com/google/jax/commit/ed849ff9e0576dcee2514741b5ffa951a94e20a8"><code>ed849ff</code></a> Make sure to call the superclass' <strong>init</strong>() on a newly created instance in P...</li> <li><a href="https://github.com/google/jax/commit/2bd1fdead81581db08ee84a0d1f82c407ccd6b11"><code>2bd1fde</code></a> Relax test tolerance in pinv test to fix a CI failure on Windows CPU.</li> <li><a href="https://github.com/google/jax/commit/e869a9d65e568e36e95940db302f94f9b7b973c4"><code>e869a9d</code></a> Merge pull request <a href="https://redirect.github.com/google/jax/issues/23415">#23415</a> from kaixih:key_value_seq_lengths</li> <li><a href="https://github.com/google/jax/commit/ea68f4569c5474f20e52b96ab88c287ab843130a"><code>ea68f45</code></a> Internal change</li> <li><a href="https://github.com/google/jax/commit/49dd6ed8d891ee6b7bbfcf7cc425382a7235556b"><code>49dd6ed</code></a> Disable a pallas export compatibility test that fails on TPU v6e.</li> <li><a href="https://github.com/google/jax/commit/808003b4e29e878349192e0f63fa1a2454ace56b"><code>808003b</code></a> Update users of jax.tree.map() to be more careful about how they handle Nones.</li> <li><a href="https://github.com/google/jax/commit/e3c4b20fa04893ad986c3184387fbd3817f1515d"><code>e3c4b20</code></a> [Pallas] Implement tiled and swizzled Memref loads for Mosaic GPU via "GPUBlo...</li> <li><a href="https://github.com/google/jax/commit/c659dc9a011bf8ff604a7e23f916920ff717288b"><code>c659dc9</code></a> [Pallas] Disable win32 gpu_ops_test.</li> <li><a href="https://github.com/google/jax/commit/14b86259d5a7fc31235f3fdfb3bcb64aadab338b"><code>14b8625</code></a> Merge pull request <a href="https://redirect.github.com/google/jax/issues/23549">#23549</a> from pschuh:docs-update</li> <li>Additional commits viewable in <a href="https://github.com/google/jax/compare/jaxlib-v0.1.32...jax-v0.4.32">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> Co-authored-by: Ilya Lavrenov <[email protected]> Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 38cf082 - Browse repository at this point
Copy the full SHA 38cf082View commit details -
Limit metrics workflow to openvino repository only (openvinotoolkit#2…
…6561) ### Details: This will prevents the metrics workflow from begin run on openvinotoolkit/openvino forks.
Configuration menu - View commit details
-
Copy full SHA for f2ea004 - Browse repository at this point
Copy the full SHA f2ea004View commit details -
[JAX FE][GHA] Add JAX/Flax Model Hub tests into GHA (openvinotoolkit#…
…26542) **Details:** Add JAX Model Hub tests into GHA **Ticket:** 142882 --------- Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for deb702e - Browse repository at this point
Copy the full SHA deb702eView commit details -
[CI] [GHA] Enable testing of Python 3.12 on Ubuntu 24 (openvinotoolki…
…t#26466) ### Tickets: - *151255* --------- Co-authored-by: Anastasia Kuporosova <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 3f0bd0f - Browse repository at this point
Copy the full SHA 3f0bd0fView commit details -
[GHA] Switch off GPU tests in precommit, post-commit and nightly (ope…
…nvinotoolkit#26559) ### Details: - Switch off GPU tests in GHA ### Tickets: - 152280
Configuration menu - View commit details
-
Copy full SHA for d5eeea1 - Browse repository at this point
Copy the full SHA d5eeea1View commit details -
[GHA] Dynamic refs for third-party OV-related repos (openvinotoolkit#…
…26564) To avoid manual updates where we can (in repos with the same branching systems as OV)
Configuration menu - View commit details
-
Copy full SHA for 2d7adc5 - Browse repository at this point
Copy the full SHA 2d7adc5View commit details -
NPUW: Introduce DQ (openvinotoolkit#26362)
### Details: - Introduced dyn quant patterns and transformations for common cases - When supported, dramatically improves performance for "Option B" when no DCOFF etc options applied - Currently experimental, disabled by default (assuming compiler will do this too), enabled by `"NPUW_DQ" : "YES"` - CW pattern works fine with prefill, GQ is generate only - not recommend to use with prefill models ### Tickets: - E-137655 (CW) - E-137657 (GQ)
Configuration menu - View commit details
-
Copy full SHA for 4b6d2c7 - Browse repository at this point
Copy the full SHA 4b6d2c7View commit details
Commits on Sep 13, 2024
-
[GPU] Fix unit tests ph3 (openvinotoolkit#26566)
### Details: - Fix errors in GPU unit tests
Configuration menu - View commit details
-
Copy full SHA for 71a7b5f - Browse repository at this point
Copy the full SHA 71a7b5fView commit details -
[GPU] Support for empty second input in Bucketize (openvinotoolkit#26292
) ### Details: - According to spec, Bucketize should support for empty second input. - In case of empty second input, the bucket index for all elements is equal to zero. - Issue is occurred when the empty second input is created as Constant. - Constant shouldn't be set by force when the user is Bucketize. ### Tickets: - 149852
Configuration menu - View commit details
-
Copy full SHA for 41b502a - Browse repository at this point
Copy the full SHA 41b502aView commit details -
[GPU] Fixes for speculative decoding (openvinotoolkit#26578)
### Details: - This PR adds additional kernel variation for proper speculative decoding handling and minor improvements of paged attention ### Tickets: - [CVS-138903](https://jira.devtools.intel.com/browse/CVS-138903)
Configuration menu - View commit details
-
Copy full SHA for 661cc03 - Browse repository at this point
Copy the full SHA 661cc03View commit details -
[GHA] Remove GenAI build from Android arm64 (openvinotoolkit#26568)
### Details: - Fix for openvinotoolkit#26546 ### Tickets: - CVS-152409
Configuration menu - View commit details
-
Copy full SHA for 577a099 - Browse repository at this point
Copy the full SHA 577a099View commit details -
[GPU] Fix broadcast save/load impls (openvinotoolkit#26582)
### Details: - Fixes sporadic crash of `KVCacheTests.smoke_multipleIterations_stateful_gather_with_initializer_cached` test ### Jira - https://jira.devtools.intel.com/browse/CVS-133748
Configuration menu - View commit details
-
Copy full SHA for c9288cc - Browse repository at this point
Copy the full SHA c9288ccView commit details -
[NPUW] Fix scalar unique names (openvinotoolkit#26541)
Currently `ov::Model` can share names for the `Constant`s even if they don't have the same name in the IR. This leads NPUW to considering the first scalar and the second one (which is related to entirely different layer) is omitted. Note, that even if OV shares the name, pointers are different. As a simple fix I'm adding output layer description to distinguish scalars in such case. TODO: follow up with a ticket to replace names in banks with pointers
Configuration menu - View commit details
-
Copy full SHA for 6783d6f - Browse repository at this point
Copy the full SHA 6783d6fView commit details -
[CI] [GHA] Use test image for test jobs (openvinotoolkit#26588)
### Tickets: - *152350*
Configuration menu - View commit details
-
Copy full SHA for 9de51dd - Browse repository at this point
Copy the full SHA 9de51ddView commit details -
[TF FE] SparseSegmentMean translator (openvinotoolkit#26540)
### Details: - Added translator for SparseSegmentMean ### Tickets: - 149705 --------- Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4e17541 - Browse repository at this point
Copy the full SHA 4e17541View commit details
Commits on Sep 14, 2024
-
[Pybind] Check version of CMAKE_CXX_COMPILER_VERSION (openvinotoolkit…
…#26550) ### Details: Python bindings compiled with the 14.28 toolset (19.28 compiler) could not be imported by another Pybind module linked with OpenVINO ### Tickets: - *149694* --------- Co-authored-by: Ilya Lavrenov <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 00861cd - Browse repository at this point
Copy the full SHA 00861cdView commit details -
[TF FE] Skip layer test for Prod on complex tensors on ARM due to acc…
…uracy issue (openvinotoolkit#26602) **Details:** Skip layer test for Prod on complex tensors on ARM due to accuracy issue **Ticket:** TBD Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 9456e8a - Browse repository at this point
Copy the full SHA 9456e8aView commit details
Commits on Sep 15, 2024
-
[GHA] Enable parallel execution of MO Python API python tests (openvi…
…notoolkit#26512) ### Details: - Enabled parallel call - Fixed tests to be compatible with pytest-xdist plugin (pass function name instead of reference) ### Tickets: - [None](openvinotoolkit#20921) Without parallel execution ![Screenshot from 2024-09-10 09-23-16](https://github.com/user-attachments/assets/927d2246-487b-47da-b428-e606aabf9cf6) With parallel execution (8 cpu cores) ![Screenshot from 2024-09-10 09-26-10](https://github.com/user-attachments/assets/93c5d4dd-147b-4fa3-a1ca-7b2c7a5415ce) Co-authored-by: Ilya Lavrenov <[email protected]> Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for d84bb3c - Browse repository at this point
Copy the full SHA d84bb3cView commit details -
[GHA] Enable parallel execution of OVC Python API python tests (openv…
…inotoolkit#26598) ### Details: - Enabled parallel call - Fixed tests to be compatible with pytest-xdist plugin (pass function name instead of reference) ### Tickets: - [None](openvinotoolkit#20922) Without parallel execution ![Screenshot from 2024-09-13 22-01-33](https://github.com/user-attachments/assets/c13a6346-cca8-4dd5-b77a-3044cc3489be) With parallel execution (8 cpu cores) ![Screenshot from 2024-09-13 22-41-36](https://github.com/user-attachments/assets/015b583f-5880-48c0-83b1-6d96cb25683d) Since I also created openvinotoolkit#26512 I thought it might be good idea to create a PR for that as well. Enjoy faster checks! ;)
Configuration menu - View commit details
-
Copy full SHA for 351e272 - Browse repository at this point
Copy the full SHA 351e272View commit details -
[DOCS] Updated json data (openvinotoolkit#26600)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for ace08e9 - Browse repository at this point
Copy the full SHA ace08e9View commit details
Commits on Sep 16, 2024
-
[Build] Update compilation options to support DPCPP + Windows config (o…
…penvinotoolkit#26320) ### Details: - Update cmake configs to use OV_COMPILER_IS_INTEL_LLVM on windows instead of older intel compiler identifier and options
Configuration menu - View commit details
-
Copy full SHA for a5e3c5e - Browse repository at this point
Copy the full SHA a5e3c5eView commit details -
[PyOV] Add release_memory to comp model (openvinotoolkit#26472)
### Details: - add release_memory method and test for it - move `generate_concat_compiled_model` and `generate_abs_compiled_model_with_data` to separate file ### Tickets: - CVS-152395
Configuration menu - View commit details
-
Copy full SHA for 15556c0 - Browse repository at this point
Copy the full SHA 15556c0View commit details -
[GPU] Swiglu OPT kernel implementation (openvinotoolkit#26172)
### Details: - Implement Swiglu OPT kernel - Use fast erf function instead of slow OpenCL erf ### Tickets: - 149484
Configuration menu - View commit details
-
Copy full SHA for 1c2cd12 - Browse repository at this point
Copy the full SHA 1c2cd12View commit details -
[TF FE]: Support complex tensors for Rsqrt operations (openvinotoolki…
…t#26247) ### Details: - Support complex tensors for `Rsqrt`operations + tests ### Tickets: - [None](openvinotoolkit#23238) --------- Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for cd64efe - Browse repository at this point
Copy the full SHA cd64efeView commit details -
Bump reviewdog/action-shellcheck from 1.26.0 to 1.27.0 (openvinotoolk…
…it#26607) Bumps [reviewdog/action-shellcheck](https://github.com/reviewdog/action-shellcheck) from 1.26.0 to 1.27.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/reviewdog/action-shellcheck/releases">reviewdog/action-shellcheck's releases</a>.</em></p> <blockquote> <h2>Release v1.27.0</h2> <h2>What's Changed</h2> <ul> <li>chore(deps): update reviewdog/reviewdog to 0.20.2 by <a href="https://github.com/github-actions"><code>@github-actions</code></a> in <a href="https://redirect.github.com/reviewdog/action-shellcheck/pull/66">reviewdog/action-shellcheck#66</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/reviewdog/action-shellcheck/compare/v1.26.0...v1.27.0">https://github.com/reviewdog/action-shellcheck/compare/v1.26.0...v1.27.0</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/reviewdog/action-shellcheck/commit/ccaafec556ffa154f112bfcb7b9c9574190b7091"><code>ccaafec</code></a> Merge pull request <a href="https://redirect.github.com/reviewdog/action-shellcheck/issues/66">#66</a> from reviewdog/depup/reviewdog/reviewdog</li> <li><a href="https://github.com/reviewdog/action-shellcheck/commit/c1ae2f62f07fb74a299bad7261e6cb6710f158ab"><code>c1ae2f6</code></a> chore(deps): update reviewdog/reviewdog to 0.20.2</li> <li>See full diff in <a href="https://github.com/reviewdog/action-shellcheck/compare/d99499e855260c9c56f7a1d066933b57326e9e7c...ccaafec556ffa154f112bfcb7b9c9574190b7091">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=reviewdog/action-shellcheck&package-manager=github_actions&previous-version=1.26.0&new-version=1.27.0)](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>
Configuration menu - View commit details
-
Copy full SHA for 96a6ece - Browse repository at this point
Copy the full SHA 96a6eceView commit details -
Bump reviewdog/action-suggester from 1.17.0 to 1.18.0 (openvinotoolki…
…t#26606) Bumps [reviewdog/action-suggester](https://github.com/reviewdog/action-suggester) from 1.17.0 to 1.18.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/reviewdog/action-suggester/releases">reviewdog/action-suggester's releases</a>.</em></p> <blockquote> <h2>Release v1.18.0</h2> <h2>What's Changed</h2> <ul> <li>chore(deps): update reviewdog/reviewdog to 0.20.2 by <a href="https://github.com/github-actions"><code>@github-actions</code></a> in <a href="https://redirect.github.com/reviewdog/action-suggester/pull/63">reviewdog/action-suggester#63</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/reviewdog/action-suggester/compare/v1.17.0...v1.18.0">https://github.com/reviewdog/action-suggester/compare/v1.17.0...v1.18.0</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/reviewdog/action-suggester/commit/db4abb16fbaabe386831e5addb7be1485d0d63d3"><code>db4abb1</code></a> Merge pull request <a href="https://redirect.github.com/reviewdog/action-suggester/issues/63">#63</a> from reviewdog/depup/reviewdog/reviewdog</li> <li><a href="https://github.com/reviewdog/action-suggester/commit/d6fc9b353f58bf7da170b0511f9aefefb8d8fcfd"><code>d6fc9b3</code></a> chore(deps): update reviewdog/reviewdog to 0.20.2</li> <li>See full diff in <a href="https://github.com/reviewdog/action-suggester/compare/63b8f8cc21dfa052ac44436e65ed31edcffcb6c1...db4abb16fbaabe386831e5addb7be1485d0d63d3">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=reviewdog/action-suggester&package-manager=github_actions&previous-version=1.17.0&new-version=1.18.0)](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>
Configuration menu - View commit details
-
Copy full SHA for 3b2c4ed - Browse repository at this point
Copy the full SHA 3b2c4edView commit details -
[Commit slider] ARM + MAC support (openvinotoolkit#26591)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 36f08f9 - Browse repository at this point
Copy the full SHA 36f08f9View commit details -
[TEST] Rerun tests if sporadic issue in tracing happens (openvinotool…
…kit#26589) ### Details: - *Rerun tests if sporadic issue in tracing happens* ### Tickets: - *CVS-151277*
Configuration menu - View commit details
-
Copy full SHA for de15414 - Browse repository at this point
Copy the full SHA de15414View commit details -
[GPU] Increasing conditions for using fake_aligned shapes (openvinoto…
…olkit#26450) ### Tickets: - *[148550](https://jira.devtools.intel.com/browse/CVS-148550)*
Configuration menu - View commit details
-
Copy full SHA for 03aaa41 - Browse repository at this point
Copy the full SHA 03aaa41View commit details -
[DOCS] Add compression note (openvinotoolkit#26558)
### Details: - *item1* - *...* ### Tickets: - CVS-150747
Configuration menu - View commit details
-
Copy full SHA for cfc0b27 - Browse repository at this point
Copy the full SHA cfc0b27View commit details -
[GHA] Support for direct package URLs specified for OV provider (open…
…vinotoolkit#26539) For passing path to release packages, for example
Configuration menu - View commit details
-
Copy full SHA for 2ea106e - Browse repository at this point
Copy the full SHA 2ea106eView commit details -
Using OV RTTI: part1 (transformations) (openvinotoolkit#26454)
### Details: Due to the issues on Android system with the standart c++ RTTI, it was proposed to switch to the ov RTTI So we are going to replace dynamic_ptr_cast to ov::as_type_ptr, dynamic_cast to ov::as_type ### Tickets: - *CVS-151640*
Configuration menu - View commit details
-
Copy full SHA for 1ccaa0f - Browse repository at this point
Copy the full SHA 1ccaa0fView commit details
Commits on Sep 17, 2024
-
[DOCS] Removing IOT Devkit references (openvinotoolkit#26620)
Removing references to internal resources of IoT Devkit found in OpenVINO documentation. This PR addresses the following JIRA ticket: 105801
Configuration menu - View commit details
-
Copy full SHA for 3349870 - Browse repository at this point
Copy the full SHA 3349870View commit details -
[DOCS] Documenting cache_encryption_callbacks option (openvinotoolkit…
…#26584) Backporting changes from: openvinotoolkit#26484 Co-authored-by: Karol Blaszczak <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 68f7d33 - Browse repository at this point
Copy the full SHA 68f7d33View commit details -
[ARM] Use OpenMP in ACL build if
THREADING=OMP
(openvinotoolkit#26619)### Details: - ACL should use OpenMP scheduler if OpenVINO is built with `THREADING=OMP` option ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 0861d86 - Browse repository at this point
Copy the full SHA 0861d86View commit details -
[DOCS] Port ov archive release version (openvinotoolkit#26622)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for fbe2b53 - Browse repository at this point
Copy the full SHA fbe2b53View commit details -
NPUW: Merge parallel MMs, if possible + Update GQ transformation (ope…
…nvinotoolkit#26580) ### Details: - When possible, do horizontal fusion for 2-3 MMs connected to the same source. - Also reword the GQ pattern for non-GPTQ models, now it is as performant as the MLIR one. ### Tickets: - E-137604 (merge) - E-138565 (DQ update)
Configuration menu - View commit details
-
Copy full SHA for 3cefe70 - Browse repository at this point
Copy the full SHA 3cefe70View commit details -
[Snippets] Introduced BufferExpression (openvinotoolkit#26413)
### Details: - *Created the first specialized expression: `BufferExpression` - now we don't have to cast the source node to `op::Buffer` to get attributes - all of them are stored in `BufferExpression` now* - *United `NewMemoryBuffer` and `IntermediateMemoryBuffer` ops into one `Buffer` since now they are distinguished only by argument count* - *Replaced the pass "SetBrgemmCopyBBuffersShape" with new "InsertBrgemmCopyBBuffers" which inserts `BrgemmCopyB`-specific `BufferExpression` after this op* ### Tickets: - *151198*
Configuration menu - View commit details
-
Copy full SHA for 0217615 - Browse repository at this point
Copy the full SHA 0217615View commit details -
[GHA] Set CC_Build timeout (openvinotoolkit#26629)
### Details: - Set timeout for CC_Build Windows job ### Tickets: - *150783*
Configuration menu - View commit details
-
Copy full SHA for ec03d9f - Browse repository at this point
Copy the full SHA ec03d9fView commit details -
[DOCS] Page adjustment (openvinotoolkit#26637)
PR for main changes with page style - sidebar resizing - H1 ,,, refactor - Scrollbar - ...
Configuration menu - View commit details
-
Copy full SHA for 220e28a - Browse repository at this point
Copy the full SHA 220e28aView commit details -
[TESTS] Use fixed seed for quantized tests (openvinotoolkit#26618)
### Details: - *Use fixed seed for quantized tests* ### Tickets: - *CVS-152410* - *CVS-138583*
Configuration menu - View commit details
-
Copy full SHA for a8bc1dc - Browse repository at this point
Copy the full SHA a8bc1dcView commit details -
[GHA] Use explicitly 2019 / 2022 toolsets (openvinotoolkit#26627)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 704ec9e - Browse repository at this point
Copy the full SHA 704ec9eView commit details -
Update jax requirement from <=0.4.32 to <=0.4.33 in /tests (openvinot…
…oolkit#26625) Updates the requirements on [jax](https://github.com/google/jax) to permit the latest version. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/google/jax/releases">jax's releases</a>.</em></p> <blockquote> <h2>JAX release v0.4.33</h2> <p>This is a patch release on top of jax 0.4.32, that fixes two bugs found in that release.</p> <p>A TPU-only data corruption bug was found in the version of libtpu pinned by JAX 0.4.32, which manifested only if multiple TPU slices were present in the same job, for example, if training on multiple v5e slices.</p> <p>This release fixes that issue by pinning a fixed version of <code>libtpu-nightly</code>.</p> <p>This release also fixes an inaccurate result for F64 tanh on CPU (<a href="https://redirect.github.com/google/jax/issues/23590">#23590</a>).</p> </blockquote> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/google/jax/blob/main/CHANGELOG.md">jax's changelog</a>.</em></p> <blockquote> <h2>jax 0.4.33 (September 16, 2024)</h2> <p>This is a patch release on top of jax 0.4.32, that fixes two bugs found in that release.</p> <p>A TPU-only data corruption bug was found in the version of libtpu pinned by JAX 0.4.32, which manifested only if multiple TPU slices were present in the same job, for example, if training on multiple v5e slices. This release fixes that issue by pinning a fixed version of <code>libtpu</code>.</p> <p>This release fixes an inaccurate result for F64 tanh on CPU (<a href="https://redirect.github.com/google/jax/issues/23590">#23590</a>).</p> <h2>jax 0.4.32 (September 11, 2024)</h2> <p>Note: This release was yanked from PyPi because of a data corruption bug on TPU. See the 0.4.33 release notes for more details.</p> <ul> <li> <p>New Functionality</p> <ul> <li>Added {func}<code>jax.extend.ffi.ffi_call</code> and {func}<code>jax.extend.ffi.ffi_lowering</code> to support the use of the new {ref}<code>ffi-tutorial</code> to interface with custom C++ and CUDA code from JAX.</li> </ul> </li> <li> <p>Changes</p> <ul> <li><code>jax_pmap_no_rank_reduction</code> flag is set to <code>True</code> by default. <ul> <li>array[0] on a pmap result now introduces a reshape (use array[0:1] instead).</li> <li>The per-shard shape (accessable via jax_array.addressable_shards or jax_array.addressable_data(0)) now has a leading (1, ...). Update code that directly accesses shards accordingly. The rank of the per-shard-shape now matches that of the global shape which is the same behavior as jit. This avoids costly reshapes when passing results from pmap into jit.</li> </ul> </li> <li><code>jax_enable_memories</code> flag is set to <code>True</code> by default.</li> <li>{mod}<code>jax.numpy</code> now supports v2023.12 of the Python Array API Standard. See {ref}<code>python-array-api</code> for more information.</li> <li>Computations on the CPU backend may now be dispatched asynchronously in more cases. Previously non-parallel computations were always dispatched synchronously. You can recover the old behavior by setting <code>jax.config.update('jax_cpu_enable_async_dispatch', False)</code>.</li> <li>Added new {func}<code>jax.process_indices</code> function to replace the <code>jax.host_ids()</code> function that was deprecated in JAX v0.2.13.</li> <li>To align with the behavior of <code>numpy.fabs</code>, <code>jax.numpy.fabs</code> has been modified to no longer support <code>complex dtypes</code>.</li> <li><code>jax.tree_util.register_dataclass</code> now checks that <code>data_fields</code> and <code>meta_fields</code> includes all dataclass fields with <code>init=True</code> and only them, if <code>nodetype</code> is a dataclass.</li> <li>Several {mod}<code>jax.numpy</code> functions now have full {class}<code>~jax.numpy.ufunc</code> interfaces, including {obj}<code>~jax.numpy.add</code>, {obj}<code>~jax.numpy.multiply</code>, {obj}<code>~jax.numpy.bitwise_and</code>, {obj}<code>~jax.numpy.bitwise_or</code>, {obj}<code>~jax.numpy.bitwise_xor</code>, {obj}<code>~jax.numpy.logical_and</code>, {obj}<code>~jax.numpy.logical_and</code>, and {obj}<code>~jax.numpy.logical_and</code>.</li> </ul> </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/google/jax/commit/80e1c94de63e7f89667cdf35f38d8fe298e97a50"><code>80e1c94</code></a> Prepare for v0.4.33 release.</li> <li><a href="https://github.com/google/jax/commit/1594d2f30fdbfebf693aba4a2b264e4a3e52acc6"><code>1594d2f</code></a> Prepare for v0.4.32 release.</li> <li><a href="https://github.com/google/jax/commit/ed849ff9e0576dcee2514741b5ffa951a94e20a8"><code>ed849ff</code></a> Make sure to call the superclass' <strong>init</strong>() on a newly created instance in P...</li> <li><a href="https://github.com/google/jax/commit/2bd1fdead81581db08ee84a0d1f82c407ccd6b11"><code>2bd1fde</code></a> Relax test tolerance in pinv test to fix a CI failure on Windows CPU.</li> <li><a href="https://github.com/google/jax/commit/e869a9d65e568e36e95940db302f94f9b7b973c4"><code>e869a9d</code></a> Merge pull request <a href="https://redirect.github.com/google/jax/issues/23415">#23415</a> from kaixih:key_value_seq_lengths</li> <li><a href="https://github.com/google/jax/commit/ea68f4569c5474f20e52b96ab88c287ab843130a"><code>ea68f45</code></a> Internal change</li> <li><a href="https://github.com/google/jax/commit/49dd6ed8d891ee6b7bbfcf7cc425382a7235556b"><code>49dd6ed</code></a> Disable a pallas export compatibility test that fails on TPU v6e.</li> <li><a href="https://github.com/google/jax/commit/808003b4e29e878349192e0f63fa1a2454ace56b"><code>808003b</code></a> Update users of jax.tree.map() to be more careful about how they handle Nones.</li> <li><a href="https://github.com/google/jax/commit/e3c4b20fa04893ad986c3184387fbd3817f1515d"><code>e3c4b20</code></a> [Pallas] Implement tiled and swizzled Memref loads for Mosaic GPU via "GPUBlo...</li> <li><a href="https://github.com/google/jax/commit/c659dc9a011bf8ff604a7e23f916920ff717288b"><code>c659dc9</code></a> [Pallas] Disable win32 gpu_ops_test.</li> <li>Additional commits viewable in <a href="https://github.com/google/jax/compare/jaxlib-v0.1.32...jax-v0.4.33">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> Co-authored-by: Ilya Lavrenov <[email protected]> Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for cee6bee - Browse repository at this point
Copy the full SHA cee6beeView commit details -
[DOCS] Fix css (openvinotoolkit#26642)
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for cd7e0e5 - Browse repository at this point
Copy the full SHA cd7e0e5View commit details
Commits on Sep 18, 2024
-
[CI] [GHA] Drop Python 3.8 testing in GHA (openvinotoolkit#26586)
### Tickets: - *151796*
Configuration menu - View commit details
-
Copy full SHA for d4da8e2 - Browse repository at this point
Copy the full SHA d4da8e2View commit details -
[NPU] Enable warnings as errors (openvinotoolkit#26557)
### Details: - enable `CMAKE_COMPILE_WARNING_AS_ERROR` for _intel\_npu_ directory (except _thirdparty_) - remove warning suppression for deprecated declarations - fix existing warnings ### Tickets: - *134706*
Configuration menu - View commit details
-
Copy full SHA for 5f86ffb - Browse repository at this point
Copy the full SHA 5f86ffbView commit details -
### Details: - *item1* - *...* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for a570074 - Browse repository at this point
Copy the full SHA a570074View commit details -
[Transformations] Extend LSTMSequenceFusion transformation with multi…
…ply-by-one pattern (openvinotoolkit#26641) **Details:** In customer model, there is a sub-graph under ShapeOf that is equivalent to multiplication by one. It can be eliminated and make LSTMSequence fusion possible **Ticket:** 149687 Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c7bc34f - Browse repository at this point
Copy the full SHA c7bc34fView commit details -
[CI] [GHA] Add JSON parsing error to errors to look for in rerunner (o…
…penvinotoolkit#26585) ### Tickets: - *151717*
Configuration menu - View commit details
-
Copy full SHA for 5bd43f1 - Browse repository at this point
Copy the full SHA 5bd43f1View commit details -
[GHA][MACOS] Fixed test dependencies to support macos x86 (openvinoto…
…olkit#26631) ### Details: - the latest available torch version for x86 - 2.2.2 ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 62309fb - Browse repository at this point
Copy the full SHA 62309fbView commit details -
Configuration menu - View commit details
-
Copy full SHA for d9d205c - Browse repository at this point
Copy the full SHA d9d205cView commit details -
[NPU] Implement extension versions properly for each method (openvino…
…toolkit#26638) ### Details: - *Implement extension versions properly for each method* ### Tickets: - *EISW-61724*
Configuration menu - View commit details
-
Copy full SHA for 77dfb1d - Browse repository at this point
Copy the full SHA 77dfb1dView commit details -
[core] Improve Constant get_vector for low precisions (openvinotoolki…
…t#26177) ### Details: - The Constant `get_vector` works correctly for low precisions. - Initialize not used bits in Constant buffer for low precisions to avoid undefined values. ### Tickets: - CVS-149867
Configuration menu - View commit details
-
Copy full SHA for b1cd3aa - Browse repository at this point
Copy the full SHA b1cd3aaView commit details -
[GHA] Actualize docker docs & add section about reproducibility (open…
…vinotoolkit#26597) Co-authored-by: Andrei Kashchikhin <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for f081f09 - Browse repository at this point
Copy the full SHA f081f09View commit details -
[NPUW]: Adding functional tests for NPUW with partitioning (openvinot…
…oolkit#25997) ### Details: - *This PR adds functional tests for NPUW launched with online partitioning, mostly the same tests that were added for unpartitioned NPUW, except for some interesting ones for folding and pipelining* - *This PR also introduces 1 accuracy test, that, however, checked on simple (in terms of computations, not structure) model for now* ### Tickets: - *ticket-id*
Configuration menu - View commit details
-
Copy full SHA for 6570afa - Browse repository at this point
Copy the full SHA 6570afaView commit details -
[GHA] Enable parallel execution of TensorFlow Lite Layer python tests (…
…openvinotoolkit#26662) ### Details: - Enabled parallel call - Fixed tests to be compatible with pytest-xdist plugin (pass function name instead of reference) - `os.path.dirname` caused saving model in `/out` directory instead of `/out/{temp_dir}` beacause it didn't have '/' char at the end and it treated `temp_dir` like a file ### Tickets: - [None](openvinotoolkit#20920) Without parallel execution ![Screenshot from 2024-09-18 14-15-16](https://github.com/user-attachments/assets/f1b00954-de59-445a-904f-5b13819c0971) With parallel execution (8 cpu cores) ![Screenshot from 2024-09-18 14-32-48](https://github.com/user-attachments/assets/2fc144cc-f771-43aa-909b-f41dedc1ccca)
Configuration menu - View commit details
-
Copy full SHA for e4dbed2 - Browse repository at this point
Copy the full SHA e4dbed2View commit details -
[DOCS] Documenting the support for MXFP4 data format. (openvinotoolki…
…t#26554) Providing info about support for MXFP4 data format in quantization on CPU. This PR addresses JIRA ticket no. 151042.
Configuration menu - View commit details
-
Copy full SHA for 1d3a711 - Browse repository at this point
Copy the full SHA 1d3a711View commit details -
Configuration menu - View commit details
-
Copy full SHA for 311e2b3 - Browse repository at this point
Copy the full SHA 311e2b3View commit details -
[JAX FE]:Support jax.lax.ge and jax.lax.gt operation for JAX (openvin…
…otoolkit#26640) ### Details: - support jax.lax.ge and jax.lax.gt operation - create unit tests ### Tickets: - [None](openvinotoolkit#26572) --------- Co-authored-by: Roman Kazantsev <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 6c9d796 - Browse repository at this point
Copy the full SHA 6c9d796View commit details -
[TF FE][SDL] Fix performance inefficiencies (openvinotoolkit#26644)
**Details:** Fix performance inefficiencies **Ticket:** 123298 Signed-off-by: Kazantsev, Roman <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 4ed5ed0 - Browse repository at this point
Copy the full SHA 4ed5ed0View commit details
Commits on Sep 19, 2024
-
[GPU] Support different input and output data type in convolution ref (…
…openvinotoolkit#26501) ### Details: - For fp model, some convolutions may not be compressed to fp16 depending on the transformation policy and those convolutions may have the fused node which is of fp16. Then convolution node input data type will be fp32 while output data type fp16. Convolution needs to support this case. ### Tickets: - 147689
Configuration menu - View commit details
-
Copy full SHA for 327f8e2 - Browse repository at this point
Copy the full SHA 327f8e2View commit details -
[GPU] Fuse more eltwises for horizontally fused FC (openvinotoolkit#2…
…6599) ### Details: - Target pattern: FCs to be fused by horizontal fusing pass and they have Add users which can be regarded as bias add. Here if we fuse the FCs as is, the fused pattern will be fused_fc -> VariadicSplit -> Add so the Adds cannot be fused to the FCs. - This PR sets such Add users as the FC's bias inputs so that the fused FC can handle them as fused bias. ### Tickets: - CVS-151841
Configuration menu - View commit details
-
Copy full SHA for bdc0110 - Browse repository at this point
Copy the full SHA bdc0110View commit details -
[Snippets] Fix sporadic issue in MHAParallelWAOptimizer (openvinotool…
…kit#26660) ### Details: - *Currently, `loops_to_split` in `MHAParallelWAOptimizer` are stored in unordered_map, so elements order is not determined. This sporadically leads to the situation when loop last iteration has work_amount more than main body's increment. This might lead to failures* - *In this PR, 'loops_to_split' are stored in vector, so loop information updates are always called for expanded loop infos in determined order: FIRST_ITER->MAIN_BODY->LAST_ITER* - *Also, the corresponding assert is added to `InsertSpecificIterations::get_decomposed_loop_work_amount` in order to throw an exception on early stage in case of incorrect configuration. This assert also allows to cover the changes by the existing tests (some of them fail if assert is added but the fix is not applied)* ### Tickets: - *N/A*
Configuration menu - View commit details
-
Copy full SHA for f0c5d2f - Browse repository at this point
Copy the full SHA f0c5d2fView commit details -
[CPU] Could not deserialize the RMS node (openvinotoolkit#26658)
### Details: - *Could not deserialize RMS node during reading model from cache* - *...* ### Tickets: - *152740*
Configuration menu - View commit details
-
Copy full SHA for 94fe66c - Browse repository at this point
Copy the full SHA 94fe66cView commit details -
[core] ShapeOf preserve lower bound when upper is infinite (openvinot…
…oolkit#21414) ### Details: - `ShapeOf` preserve lower bound when upper is infinite ### Tickets: - [CVS-126430](https://jira.devtools.intel.com/browse/CVS-126430)
Configuration menu - View commit details
-
Copy full SHA for 67a4c18 - Browse repository at this point
Copy the full SHA 67a4c18View commit details -
[GPU]: Fix for failing i16 elementwise functional test. (openvinotool…
…kit#26383) ### Details: - Fix for failing GPU functional i16 test. The problem is that i16 input is wrongly converted to f32 in constant and parameter op. - Had to disable i16 case for Deformable conv, which won't work with this fix. THe motivation is that Deformable conv on GPU supports only f16, f32 and int8 types - does not support i16 case, which was working only due to some implicit type transformation which this PR changes.
Configuration menu - View commit details
-
Copy full SHA for 60325de - Browse repository at this point
Copy the full SHA 60325deView commit details -
[OV JS] Prepare testing and codestyle guidelines (openvinotoolkit#26616)
### Details: - Add guidelines how to test new js api functionality - Add guide how to extend JS API functionality ### Tickets: - [CVS-151489](https://jira.devtools.intel.com/browse/CVS-151489) [CVS-151492](https://jira.devtools.intel.com/browse/CVS-151492) --------- Co-authored-by: Tatiana Savina <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for b368c31 - Browse repository at this point
Copy the full SHA b368c31View commit details -
[DOCS] Port benchmark 24.4 (openvinotoolkit#26690)
### Details: - *item1* - *...* ### Tickets: - *ticket-id* Co-authored-by: Karol Blaszczak <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 76fab4c - Browse repository at this point
Copy the full SHA 76fab4cView commit details -
[CI] [GHA] Add
lost communication with the server
to errors to look…… for in rerunner (openvinotoolkit#26691) ### Tickets: - *152565*
Configuration menu - View commit details
-
Copy full SHA for 42b7322 - Browse repository at this point
Copy the full SHA 42b7322View commit details
Commits on Sep 20, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 2b84653 - Browse repository at this point
Copy the full SHA 2b84653View commit details