### Details:
Numpy started preparation to release 2.0 version.
It brings many breaking changes listed below.
Align with these changes it's a plan for further development
(CVS-138766), for today we want to secure users against unintentional
installation of 2.x version of numpy with OV.
This PR has to be ported for 2024/1 branch
- Preparing to restrict OV to not use release of Numpy 2.0 which may
introduce many breaking changes including:
- required bump of `pybind` to 2.12+
- API and namespaces clean-ups and deprecations
- behavior changes of keywords such as `copy`
- changes to existing and introduction of new dtypes (such as
`StringDType` and possibility of adding custom ones)
- more can be found here:
https://numpy.org/devdocs/release/2.0.0-notes.html
- and here:
https://numpy.org/devdocs/numpy_2_0_migration_guide.html#numpy-2-migration-guide
### Tickets:
- CVS-138838
---------
Co-authored-by: Michal Lukaszewski <michal.lukaszewski@intel.com>
### Details:
- *[API][AUTO] Plugin is not supported some mandatory hint properties*
### Tickets:
- *CVS-137446, CVS-136746, CVS-136754*
---------
Co-authored-by: Chen Peter <peter.chen@intel.com>
### Details:
- Correct tensor shape of AUTO before setting it to selected HW infer
request.
- Remove CPU shape.compatible checks when setting output tensor with
dynamic shape
### Tickets:
- CVS-136756
### Details:
- Constant's function `get_data` is extended to provide both copy and
view of the data.
- `get_data` can now cast date to desired dtype.
- Added testcases.
Example:
```python
In [1]: import openvino as ov
In [2]: import numpy as np
In [3]: arr = np.array([1.0, 1012.5, 0.0, 0.1, 0.5], dtype=np.float32)
In [4]: c = ov.runtime.opset13.constant(arr, dtype=np.float32)
In [5]: c.get_data(dtype=np.bool_)
Out[5]: array([ True, True, False, True, True])
In [6]: c = ov.runtime.opset13.constant(arr, dtype=ov.Type.bf16)
In [7]: c.get_data(dtype=np.float32)
Out[7]:
array([1.0000000e+00, 1.0120000e+03, 0.0000000e+00, 9.9609375e-02,
5.0000000e-01], dtype=float32)
In [8]: c.get_data(dtype=np.float32, copy=False)
Out[8]:
array([1.0000000e+00, 1.0120000e+03, 0.0000000e+00, 9.9609375e-02,
5.0000000e-01], dtype=float32)
```
### Tickets:
- CVS-129930
- CVS-127292
### Details:
- Fix wrong binding to 6E+1P cores on MTL windows with command "-pin
HYBRID_AWARE -nthreads 7"
- Correct binding cores: 6P+1E
- Add test case
- Add CPU debug on Windows and Mac
### Tickets:
- *CVS-138630*
### Details:
- Currently, in dynamic shape, ofm size is fixed as 2 because it is is
tightly coupled by weight layout (i.e., tile_ofm == 2 requires
os_iyx_osv32 weight format)
- For Vec*Mat case (b==1, used for 2nd+ token generation in LLMs) with
small output size N, this configuration is not good because of the
limited # of gpu threads.
- To increase GPU threads, needed to 1) reduce the ofm size to 1 2) But
still share the weight format with first token case where ofm=2 is good.
- Previously the os_iyx_osv32 weight format was packing int4 weight with
the following order:
k0_f0f16 | k0_f1f17 | .... | k0_f15f31 || k1_f0f16 | k1_f1f17 | ... |
k1_f15f31
- To use common weight format for ofm=1 and ofm=2, implemented a new
weight format for int4 dtype, where the weight is stored int the
following order:
f0_k0k1 | f1_k0k1 | .... | f15_k0k1|| f16_k0k1 | f17_k0k1 | ... |
f31_k0k1
### Tickets:
- 138296
### Details:
- Removed internal conversion of all typed passed to `convert_model()`
to `np.dtype`, as some types do not have `numpy` analogs, which causes
errors. Also this step is redundant as we can use `ov.Type` instead.
### Tickets:
- 138755
### Details:
Changes in intel-npu plugin:
- Removing ov::internal properties from public supported_properties list
- publishing model_priority property in supported_properties
**Details:** Xfail Mod and FloorDiv tests due to sporadic inference
mismatch results. Correct Conv3D test to decrease workload to pass
timeout.
**Tickets:** 138904, 138713
Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
### Details:
- use custom compressed action to restore/save cache, it should solve
network related issued with sccache and speedup the build from cache
(from 10 to 6 minutes)
-
### Tickets:
- *ticket-id*
### Details:
- add `Core.import_model(stream, device_name, config[optional})
- add ` ov::AnyMap to_anyMap(const Napi::Env&, const Napi::Value&)`
helper for conversion from Napi::Value to ov::AnyMap and reuse it in
other methods
### Tickets:
- *136460*
HETERO plugin ('HETERO:GPU,CPU') raises an error [GPU] Mismatch tensor
and port types: i64 vs i32
### Details:
- *Do not convert Result element type after Extension if required;*
- *Eliminate converts after the second call of the ConvertPrecisoin
pass;*
- *Support string input in the `benchmark_app`*.
### Tickets:
- *136752*
### Details:
- This PR introduces a PR labeller for external PRs. It will add the
`ExternalPR` label to a PR whose author is not in the `openvino`
collaborator list.
It will not work from a PR as it has a `pull_request_target` trigger for
the secrets. I've tested it locally.
### Tickets:
- *135090*
### Details:
- Limit NF4 conversion to FP16 -> NF4 in convert operator as values are
correctly quantized, Not support NF4 conversion to/from other types.
### Tickets:
- [CVS-135304](https://jira.devtools.intel.com/browse/CVS-135304)
This fixes#23566
Implemented Model.setFriendlyName and Model.getFriendlyName in Node.js
api
- Added Parameter Validation
- Added Tests in model.test.js
---------
Co-authored-by: Vishniakov Nikolai <nikolai.vishniakov@intel.com>
This Fixes#23441 issue.
Details:
1. Modified get_shape() method to report error in case any argument is
passed.
2. Added Required Tests for the above in `tensor.test.js`.
---------
Co-authored-by: Vishniakov Nikolai <nikolai.vishniakov@intel.com>
Co-authored-by: Alicja Miloszewska <alicja.miloszewska@intel.com>
### Details:
- The compilation of a model without inputs was failing because the
serialization in zero compiler expected at least one input.
### Tickets:
- EISW-107905
Ivan Novoselov