Resolve: [Good First Issue][Keras 3 OpenVINO Backend]: Support numpy.linspace operation #29485

[VLTisME]

Reduced failed tests to 2 failed and 12 skipped.
two left unsolved failed tests:
1/
start = 0, stop = 10, num = <tf.Tensor: shape=(), dtype=int32, numpy=5>
endpoint = True, retstep = False, dtype = None, axis = 0

@keras_export(["keras.ops.linspace", "keras.ops.numpy.linspace"])
def linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None, axis=0, *, device=None):
    # Validate num
    if not isinstance(num, int):

      raise TypeError("num must be an integer")

E TypeError: num must be an integer

2/

  arrays = [np.empty(x, dtype=[]) for x in args]

E TypeError: 'NoneType' object cannot be interpreted as an integer

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[VLTisME]

@rkazants may you review this?

[VLTisME]

I have collected some errors, gonna fix it:
1/

      start_np, stop_np = np.broadcast_arrays(start.numpy(), stop.numpy())

E AttributeError: 'OpenVINOKerasTensor' object has no attribute 'numpy'
2/

      start_np, stop_np = np.broadcast_arrays(start.numpy(), stop.numpy())

E AttributeError: 'numpy.ndarray' object has no attribute 'numpy'. Did you mean: 'dump'?
3/
if backend.shape(start) == () and backend.shape(stop) == ():
result = start + indices * step
else:

      start_np, stop_np = np.broadcast_arrays(start.numpy(), stop.numpy())

E AttributeError: 'jaxlib.xla_extension.ArrayImpl' object has no attribute 'numpy'
4/
def linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None, axis=0, *, device=None):
if not isinstance(num, int):

      raise TypeError("num must be an integer")

E TypeError: num must be an integer
5/

arrays = [np.empty(x, dtype=[]) for x in args]
E TypeError: 'NoneType' object cannot be interpreted as an integer

[rkazants]

incorrect file is modified. You need to update https://github.com/keras-team/keras/blob/master/keras/src/backend/openvino/numpy.py

[rkazants]

please remove this file. It is needed only for developer purpose

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 0% with 74 lines in your changes missing coverage. Please review.

Project coverage is 79.83%. Comparing base (99a6bdc) to head (c4e8245).
Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
keras/src/backend/openvino/numpy.py	0.00%	74 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master   #21140      +/-   ##
==========================================
- Coverage   82.59%   79.83%   -2.76%     
==========================================
  Files         564      564              
  Lines       54367    54440      +73     
  Branches     8452     8466      +14     
==========================================
- Hits        44902    43460    -1442     
- Misses       7391     8966    +1575     
+ Partials     2074     2014      -60     

Flag	Coverage Δ
keras	79.64% <0.00%> (-2.76%)	⬇️
keras-jax	63.69% <0.00%> (-0.09%)	⬇️
keras-numpy	58.80% <0.00%> (-0.08%)	⬇️
keras-openvino	?
keras-tensorflow	64.09% <0.00%> (-0.09%)	⬇️
keras-torch	63.77% <0.00%> (-0.09%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[rkazants]

@VLTisME, please fix test failures, see CI logs:

FAILED keras/src/ops/numpy_test.py::NumpyTwoInputOpsCorrectnessTest::test_linspace - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'
FAILED keras/src/ops/numpy_test.py::NumpyDtypeTest::test_linspace_[0, 10]_0_bfloat16 - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'
FAILED keras/src/ops/numpy_test.py::NumpyDtypeTest::test_linspace_[0, 10]_0_float16 - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'
FAILED keras/src/ops/numpy_test.py::NumpyDtypeTest::test_linspace_[0, 10]_0_float32 - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'
FAILED keras/src/ops/numpy_test.py::NumpyDtypeTest::test_linspace_[0, 10]_0_float64 - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'
FAILED keras/src/ops/numpy_test.py::NumpyDtypeTest::test_linspace_[0, 10]_0_none - TypeError: unsupported operand type(s) for -: 'openvino._pyopenvino.Output' and 'openvino._pyopenvino.Output'

[VLTisME]

i am working on it, thanks for your review

[VLTisME]

Why do all the tests pass when I run them locally (except for a few skipped ones), but when I push the code here, so many tests fail?
P/S: I run locally by first creating a pytest.ini and write the content in there, then:

pytest-c ./pytest.ini ./keras/src/ops/numpy_test.py