f6e6f2af56
[CPU][ARM] Support deconvolution to correctly handle multiple output edges on a single output port ( #24754 )
2024-06-07 17:32:26 +00:00
45147aefa0
[GHA] dGPU tests pipeline ( #24572 )
...
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
2024-06-07 15:22:07 +00:00
beb37eb4e1
[GPU] Move scaling from QK dot product to Q ( #24896 )
...
### Details:
- Move scaling from QK dot product to Q input for better SDPA accuracy
with high-value ranges
### Tickets:
- [CVS-143256](https://jira.devtools.intel.com/browse/CVS-143256 )
2024-06-07 13:04:12 +00:00
d996280d5e
[CPU] Cherry-pick IP optimizations from latest oneDNN master ( #24654 )
...
### Details:
- Downstream latest InnerProduct performance optimizations to oneDNN
fork
- Ticket:
[CVS-142995](https://jira.devtools.intel.com/browse/CVS-142995 )
- oneDNN fork PR: https://github.com/openvinotoolkit/oneDNN/pull/252
2024-06-07 12:53:38 +00:00
2893f2fc7c
[Snippets] Added dynamism support to MHA Tokenization pass ( #24897 )
...
### Details:
- *Updated `MHATokenization` pass to support dynamic nodes tokenization*
- *Added MHATokenization unit tests*
- *Disabled dynamic MHA tokenization in CPU Plugin*
### Tickets:
- *123329*
2024-06-07 12:13:26 +00:00
ba5c45a4ee
[PyOV] Fix hanging on infer request destruction ( #24722 )
...
### Details:
- Initial problem: `test_custom_op` hanged on destruction because it was
waiting for a thread which tried to acquire GIL.
- The second problem is that pybind11 doesn't allow to work with GIL
besides of current scope and it's impossible to release GIL for
destructors. https://github.com/pybind/pybind11/issues/1446
- Current solution allows to release GIL for InferRequest and all called
by chain destructors.
### Tickets:
- CVS-141744
2024-06-07 11:56:36 +00:00
bb179c69d1
[DOCS] CPU perf hints doc review ( #24827 )
...
### Details:
- *item1*
- *...*
### Tickets:
- *ticket-id*
2024-06-07 11:46:26 +00:00
c542405470
More accurate shape of optimization ( #24845 )
...
### Details:
- *Topological order was corrected to perform the best ShapeOf
reconnections possible*
### Tickets:
- *CVS-142953*
2024-06-07 10:34:53 +00:00
5faf95018d
[DOCS] fix back to top button ( #24884 )
2024-06-07 10:28:41 +00:00
852878162d
[TRANSFORMATIONS] Add a check to SDPAToPagedAttention to fail fast ( #24841 )
...
[TRANSFORMATIONS] Add a check to SDPAToPagedAttention to fail fast
Add a check to SDPAToPagedAttention to fail fast if no
ScaledDotProductAttention operation present in a graph hence no
transformation will be executed.
### Tickets:
- CVS-143067
Signed-off-by: Andrii Staikov <andrii.staikov@intel.com>
---------
Signed-off-by: Andrii Staikov <andrii.staikov@intel.com>
Co-authored-by: Ivan Tikhonov <ivan.tikhonov@intel.com>
2024-06-07 08:23:27 +00:00
7f1ddd55ac
[Core/Ref] Resolve coverity issues ( #24874 )
...
### Details:
- Fixed coverity issues in src/core/reference
### Tickets:
- CVS-143152
2024-06-07 08:14:32 +00:00
236e1062b2
[CPU] Optimize the unique operator ( #24850 )
...
### Details:
- *Optimize unique using HashMap when sorted=false*
- *...*
### Tickets:
- *ticket-id*
2024-06-07 07:06:10 +00:00
d286268a85
[GPU][DOC] Link to OCL article ( #24813 )
2024-06-07 06:38:58 +00:00
faa19e6121
[GPU] Fix wrong condition for dynamic onednn convolution impl-type. ( #24497 )
...
Remove the limitation that oneDNN convolution doesn't support dynamic
input and output.
With this limitation, clDNN convolution will run forever in dpas
platform.
Because oneDNN Opt convolutions require blocked-format and current shape
agnostic convolution doesn't support block-format, convolution do
nothing in compile_graph stage and actual kernel compilation will happen
at first inference.
### Tickets:
- *141040*
Signed-off-by: hyunback <hyunback.kim@intel.com>
2024-06-07 06:32:38 +00:00
00510e69fd
[Snippets] Fixed Invalid Read in FuseLoops pass ( #24880 )
...
### Details:
- *Fixed Invalid Read in FuseLoops pass. Before we save loop ports by
reference from `current_loop_info`. However, this `current_loop_Info`
might be changed (another shared pointer) and the previous saved loop
ports by reference are invalid (the original `current_loop_info` is
expired)*
- *`Valgrind` doesn't throw exception anymore with this fix*
### Tickets:
- *N/A*
2024-06-07 06:29:01 +00:00
416c4ffb7e
[CORE] depracate get/set_concatenation_axis functions ( #24737 )
...
### Details:
- remove `set_concatenation_axis` function usage
- replace `get_concatenation_axis` function usage
### Tickets:
- [CVS-94507](https://jira.devtools.intel.com/browse/CVS-94507 )
### Related PR:
- https://github.com/openvinotoolkit/openvino/pull/24383
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2024-06-07 05:55:52 +00:00
0cc051debb
[GPU] Don't add crop for unused out ports of Split op ( #24877 )
...
### Details:
- Currently we insert crop primitive for each out port of
(Variadic)Split ops, and if some of them are not used we just waste some
execution time. This patch modifies converter for Split to process ports
with real users only.
2024-06-07 05:53:13 +00:00
b8ced3dfc7
Change opt-125m-gptq model ( #24899 )
...
### Details:
- *item1*
- *...*
### Tickets:
- *ticket-id*
2024-06-07 14:50:30 +04:00
4fcfaf24b3
Bump actions/dependency-review-action from 4.3.2 to 4.3.3 ( #24876 )
...
Bumps
[actions/dependency-review-action](https://github.com/actions/dependency-review-action )
from 4.3.2 to 4.3.3.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/dependency-review-action/releases ">actions/dependency-review-action's
releases</a>.</em></p>
<blockquote>
<h2>Notes for v4.3.3</h2>
<h2>What's Changed</h2>
<ul>
<li>Allow slashes in purl package names by <a
href="https://github.com/juxtin "><code>@juxtin</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/765 ">actions/dependency-review-action#765</a></li>
<li>use the v3 version of the deps.dev API by <a
href="https://github.com/josieang "><code>@josieang</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/741 ">actions/dependency-review-action#741</a></li>
<li>PR with suggestions - [Improvement]: Help streamline / simplify
dependency review action README by <a
href="https://github.com/am-stead "><code>@am-stead</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/773 ">actions/dependency-review-action#773</a></li>
<li>fix show-openssf-scorecard-levels input by <a
href="https://github.com/ramann "><code>@ramann</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/776 ">actions/dependency-review-action#776</a></li>
<li>Updates to the contribution guidelines by <a
href="https://github.com/jonjanego "><code>@jonjanego</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/778 ">actions/dependency-review-action#778</a></li>
<li>Create issue templates by <a
href="https://github.com/jonjanego "><code>@jonjanego</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/777 ">actions/dependency-review-action#777</a></li>
<li>Fix the max comment length issue by <a
href="https://github.com/jhutchings1 "><code>@jhutchings1</code></a> and
<a href="https://github.com/elireisman "><code>@elireisman</code></a> in
<a
href="https://redirect.github.com/actions/dependency-review-action/pull/767 ">actions/dependency-review-action#767</a></li>
<li>Bump project version to 4.3.3 in prep for a release by <a
href="https://github.com/elireisman "><code>@elireisman</code></a> in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/781 ">actions/dependency-review-action#781</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a href="https://github.com/josieang "><code>@josieang</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/741 ">actions/dependency-review-action#741</a></li>
<li><a href="https://github.com/am-stead "><code>@am-stead</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/773 ">actions/dependency-review-action#773</a></li>
<li><a href="https://github.com/ramann "><code>@ramann</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/dependency-review-action/pull/776 ">actions/dependency-review-action#776</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/dependency-review-action/compare/v4.3.2...v4.3.3 ">https://github.com/actions/dependency-review-action/compare/v4.3.2...v4.3.3 </a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="72eb03d02chttps://redirect.github.com/actions/dependency-review-action/issues/781 ">#781</a>
from actions/release-v4.3.3</li>
<li><a
href="137d8b42cee6b618ed05https://redirect.github.com/actions/dependency-review-action/issues/767 ">#767</a>
from actions/max-comment-length</li>
<li><a
href="3c426492048e6ea8d29b1b3d2772d0220872c81a087d0f81a54531204be7df1ca890c50c155c5e85...72eb03d02chttps://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/dependency-review-action&package-manager=github_actions&previous-version=4.3.2&new-version=4.3.3 )](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] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-06-06 07:22:33 +00:00
b9e1ef7486
[DOCS] Updating Nodejs API with descriptions ( #24866 )
...
Adding descriptions of methods to Node.js API.
2024-06-06 06:58:31 +00:00
11ad74dcda
[Core/test] Resolve TODOs in chech_opset1 ( #24819 )
...
### Details:
- Removed potential tests of operator which won't be added to opset 1
(as it's frozen),
- enabled those which are present.
- Renamed function.
### Tickets:
- n/a
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2024-06-06 05:02:37 +00:00
e922edf299
Add test case to verify 4bit precision input model. ( #24774 )
...
### Details:
- *Complete the remaining work after the
[PR](https://github.com/openvinotoolkit/openvino/pull/24297 ) is merged.*
- *Add test case.*
### Tickets:
- *141542*
---------
Signed-off-by: xipingya <xiping.yan@intel.com>
Co-authored-by: Zlobin Vladimir <vladimir.zlobin@intel.com>
2024-06-06 03:23:26 +00:00
b1bc59e72d
[GPU] Fix Loop to cover the case where memory is null at the first iteration ( #24776 )
...
### Details:
- Handle the case where from_mem is null at the first iteration
### Tickets:
- 141652
2024-06-05 21:49:34 +00:00
9d208241f0
[intel-npu] Adding new properties to tests ( #24865 )
...
### Details:
- Adding missing property tests to npuFuncTests: EXECUTION_MODE_HINT,
DEVICE_TYPE, PCI_INFO and DEVICE_GOPS
### Tickets:
- none
2024-06-05 18:48:13 +00:00
a156b69ccc
[Core] Deprecate ambiguous ov::element::Type ctor ( #24698 )
...
### Details:
- Added deprecation mark to ambiguous ov::element::Type ctor
(https://github.com/openvinotoolkit/openvino/pull/11211#discussion_r834939746 )
### Tickets:
- CVS-133840
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2024-06-05 18:30:41 +00:00
00d525f5a2
Export OpenVINOGenAI_DIR ( #24683 ) ( #24870 )
...
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
2024-06-05 22:43:26 +04:00
13922e7515
[GHA] Win-CC decrease build jobs ( #24869 )
...
### Details:
Without thread limitation win-cc pipeline fails with run out of memory
error:
https://github.com/openvinotoolkit/openvino/actions/runs/9384908136
set 16 threads
### Tickets:
- *ticket-id*
2024-06-05 17:10:04 +00:00
c87cf86489
[ONNX] Align behavior of ONNX Frontend operator ReduceMean-11, 13, 18 with original framework ( #23148 )
...
### Details:
- Align behavior of ONNX Frontend operator ReduceMean-11, 13, 18 with
original framework
### Tickets:
- Closes https://github.com/openvinotoolkit/openvino/issues/20556
---------
Co-authored-by: Georgy Krivoruchko <georgy.krivoruchko@intel.com>
2024-06-05 13:30:37 +00:00
de4d00a597
[Core] Fix coverity issues ( #24838 )
...
### Details:
- Used move instead of copy.
### Tickets:
- CVS-121621
2024-06-05 12:12:57 +00:00
41d460f7ac
[GHA] Pinned MSVC toolset version ( #24860 )
...
### Details:
- Pinned MSVC toolset version
- removed hardcoded jobs count
### Tickets:
- *ticket-id*
2024-06-05 12:10:14 +00:00
9e662351e8
torch.compile documentation updates for 2024.2 ( #24803 )
...
### Details:
- Updated 'How to Use' and 'Options' sections
- Added quantization and torchserve sections
---------
Co-authored-by: Mustafa Cavus <mustafa.cavus@intel.com>
Co-authored-by: suryasidd <surya.siddharth.pemmaraju@intel.com>
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
Co-authored-by: Karol Blaszczak <karol.blaszczak@intel.com>
2024-06-05 08:33:04 +00:00
aa04a9bcfa
[WIN-CC] Support the latest msvc (19.40) ( #24849 )
...
### Details:
- Aligned build jobs with runner configuration to speed up the build
- added temporary variable data_type to resolve internal compilation
error with the new msvc (19.40)
https://jira.devtools.intel.com/browse/CVS-142993
### Tickets:
- *ticket-id*
2024-06-05 08:32:25 +00:00
cf22524dcb
[OV JS] Add methods descriptions for Model, Tensor & InferRequest ( #24768 )
...
### Details:
- Add methods descriptions for Model, Tensor & InferRequest
---------
Co-authored-by: Maciej Smyk <maciejx.smyk@intel.com>
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
2024-06-05 07:37:00 +00:00
dd0846b8d5
[CPU] [Snippets] Fix performance regression in calling parallel_nt ( #24832 )
...
### Details:
- *This PR fixes geomean level performance issue caused by
https://github.com/openvinotoolkit/openvino/pull/23180 on platforms with
sufficient cores.*
- *In the PR mensioned above, `parallel_nt` is used for executing
subgraph instead of `parallel_for5d`, becuase the former has performance
advantage for the case where #tasks is more than #threads*. That is
becuase using `parallel_nt`, we can initilize parameters for each
thread, while parameters are initilized for each task in using
`parallel_for5d`. So `parallel_nt` has less overheads.
- *However, for the case where #tasks is less than #threads,
`parallel_nt` brings performance regression. Because compared with
`parallel_for5d`, `parallel_nt` doesn't apply the argument
`tbb::static_partitioner{}` when calling `tbb::parallel_for`. Here we
use `parallel_nt_static` instead to fix the performance regression.*
2024-06-05 07:27:36 +00:00
f952f73d6b
[GPU] Fix decompression scales reorder when data is reused ( #24842 )
...
### Details:
- In case when scale tensor was used by multiple compressed FCs we had
an exception on attempt to insert reorder between them. This patch
changes reorder name so now multiple reorders are inserted for each
Scale+FC pair, and then on later stages we merge them if they do same
tensor transform.
2024-06-05 06:10:37 +00:00
adb590d5f4
[GPU] Added RoPE support for Llama2, Llama3 and Phi-3 ( #24829 )
...
### Tickets:
- *[119150](https://jira.devtools.intel.com/browse/CVS-119150 )*
- *[141777](https://jira.devtools.intel.com/browse/CVS-141777 )*
2024-06-04 22:16:48 +00:00
c45a46edca
[TFLite] Added implementation details ( #24835 )
...
### Details:
- Added an implementation details for sparse-tensors reading
### Tickets:
- N/A
---------
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
2024-06-04 18:49:05 +00:00
52d0a44d3f
[CI] [GHA] Set actions references to specific commit hash; provide explicit top level token permissions for jobs ( #24839 )
...
### Details:
- Token permissions docs:
98ec491a88/docs/checks.md (token-permissions)98ec491a88/docs/checks.md (pinned-dependencies)
2024-06-04 18:17:51 +00:00
65b5b8e1b7
[MOC] Implement EliminateLoopInputsOutputs ( #24786 )
...
**Details:** EliminateLoopInputsOutputs transformation helps to avoid an
isolated sub-graphs `Parameter->Result` in a body graph of `Loop` and
`TensorIterator`. Such sub-graphs are useless and can by-pass `Loop`
operation.
**Ticket:** 142440
---------
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: Evgeny Kotov <evgeny.kotov@intel.com>
Co-authored-by: Tikhonov Ivan <ivan.tikhonov@intel.com>
2024-06-04 12:37:16 +00:00
1a70c4002c
[GPU] Optimize RMS kernel ( #24831 )
...
### Details:
- Optimize RMS kernel by subgroup block IO
- Compute maximum possible LWS and the number of data in subgroups
- Add test cases which covers leftover handling and unaligned data shape
### Tickets:
- 141845
---------
Signed-off-by: Andrew Park <andrew.park@intel.com>
2024-06-04 11:02:12 +00:00
d954a80161
[CPU] Add inplace and jit for ROPE ( #24788 )
...
### Details:
- *Add inplace and jit if possible*
- *...*
### Tickets:
- *ticket-id*
2024-06-04 10:20:18 +00:00
a23c9d5b5a
[DOCS] benchmark data for LLM master ( #24834 )
2024-06-04 09:25:38 +00:00
1a6ddfff60
[NPU] Add the "create_executor" property inside the Plugin's list as well ( #24828 )
...
### Details:
- `ov::intel_npu::create_executor` is declared a supported property only
for the `intel_npu::CompiledModel` class. The current PR adds an entry
inside `intel_npu::Plugin` as well.
### Tickets:
- *CVS-143060*
2024-06-04 09:19:21 +00:00
f1862e4d20
build: fix Python platform architecture tag for armv7l ( #24772 )
...
On arm32 (armv7), the platform tag should be `armv7l` as per
https://peps.python.org/pep-0599/
2024-06-04 12:54:20 +04:00
df6a25800d
[GPU] Added RoPE support for ChatGLM and Qwen ( #24756 )
...
### Details:
- Added support RoPE for ChatGLM and Qwen models
- Moved and refactored RoPE functional tests
### Tickets:
- *[119150](https://jira.devtools.intel.com/browse/CVS-119150 )*
2024-06-03 10:34:13 +00:00
ba8d6c5de5
[GPU] SDPA minor improvements ( #24770 )
...
### Details:
- Change the default softmax accumulator type from FP16 to FP32 (this
doesn't affect performance, as well as accuracy at first glance;
however, let's stick to higher precision here)
- Applied review comments from PR
https://github.com/openvinotoolkit/openvino/pull/24466
- Minor internal operations refactoring (SDPA and IndirectSDPA)
- Add new tests for GQA optimization and for indirect inputs
2024-06-03 10:34:13 +00:00
cea8e1f5e4
[GPU] Fixed unexpected kv cache optimization turned off ( #24805 )
...
### Details:
- kv cache opt was disabled due to the wrongly detected beam table size.
Fixed this issue.
### Tickets:
- *ticket-id*
2024-06-03 10:34:13 +00:00
142114ccbc
[Op] Squeeze update - no throw on non-squeezable dims ( #24715 )
...
### Details:
- Changes needed to enable Squeeze (and models) with axes pointing to
non-squeezable dims (currently ov throws an error for such case) -
considered as backward compatible change
- If agreed to update functionality within the same version of the op,
actual changes are needed only in shape infer related functions to cover
**Template, CPU and GPU plugins**:
- src/core/shape_inference/include/squeeze_shape_inference.hpp
- src/plugins/intel_cpu/src/shape_inference/custom/reshape.cpp
- The rest of the changes are test related.
- Pytorch FE Squeeze test enabled - passed
### Tickets:
- 142387
2024-06-03 09:59:07 +00:00
56d184dbb1
[OV JS] Add method descriptions for Core ( #24693 )
...
### Details:
- Add method descriptions to TS for Core
### Tickets:
- *141290*
---------
Co-authored-by: Sebastian Golebiewski <sebastianx.golebiewski@intel.com>
2024-06-03 07:31:07 +00:00
f859992efe
[Tests] Fixed debug info about ranges ( #24790 )
...
### Details:
- *Removed Debug info printing by default from
https://github.com/openvinotoolkit/openvino/pull/23428 *
```
[ RUN ]
smoke_Snippets_Softmax/Softmax.CompareWithRefImpl/IS=[?.?]_TS=((1.16))_((1.32))_((1.1))_((1.9))_((1.17))_((1.19))_((1.49))_((1.50))_((5.16))_((1.16))_((1.9))_Axis=-1_#N=1_#S=1_targetDevice=CPU
MEM_USAGE=3007144KB
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
RANGE FOR PARAMETER: Parameter_1590025 start from: 0.000000 range: 8
resolution: 32 seed: 1
[ OK ]
smoke_Snippets_Softmax/Softmax.CompareWithRefImpl/IS=[?.?]_TS=((1.16))_((1.32))_((1.1))_((1.9))_((1.17))_((1.19))_((1.49))_((1.50))_((5.16))_((1.16))_((1.9))_Axis=-1_#N=1_#S=1_targetDevice=CPU
(132 ms)
```
### Tickets:
- *N/A*
2024-06-03 10:56:36 +02:00
Nesterov Alexander