Fix Implicit ALS matrix zero assignment size on GPU

[chezou]

Description

Fix ValueError: YtY and Y should have the same number of columns on ImplicitALSWrapper with GPU.

Before

➜  RecTools git:(fix-als-gpu) poetry run pytest tests/models/test_implicit_als.py
================================================= test session starts ==================================================
platform linux -- Python 3.12.3, pytest-8.3.3, pluggy-1.5.0
rootdir: /home/aki/src/RecTools
configfile: setup.cfg
plugins: cov-5.0.0, subtests-0.12.1, typeguard-4.4.1, mock-3.14.0
collected 77 items

tests/models/test_implicit_als.py .....................F.......................................................  [100%]

======================================================= FAILURES =======================================================
____________________ TestImplicitALSWrapperModel.test_happy_path_with_features[True-expected0-True] ____________________

self = <tests.models.test_implicit_als.TestImplicitALSWrapperModel object at 0x7fa8d0655af0>
fit_features_together = True
expected =   user_id item_id  rank
0      u1      i2     1
1      u3      i3     1
2      u3      i2     2
use_gpu = True
dataset_w_features = Dataset(user_id_map=IdMap(external_ids=array(['u1', 'u2', 'u3'], dtype=object)), item_id_map=IdMap(external_ids=array(...tored elements in Compressed Sparse Row format>, names=(('f1', '__is_direct_feature'), ('f2', '__is_direct_feature'))))

    @pytest.mark.parametrize(
        "fit_features_together,expected",
        (
            (
                True,
                pd.DataFrame(
                    {
                        Columns.User: ["u1", "u3", "u3"],
                        Columns.Item: ["i2", "i3", "i2"],
                        Columns.Rank: [1, 1, 2],
                    }
                ),
            ),
            (
                False,
                pd.DataFrame(
                    {
                        Columns.User: ["u1", "u3", "u3"],
                        Columns.Item: ["i2", "i2", "i3"],
                        Columns.Rank: [1, 1, 2],
                    }
                ),
            ),
        ),
    )
    def test_happy_path_with_features(
        self, fit_features_together: bool, expected: pd.DataFrame, use_gpu: bool, dataset_w_features: Dataset
    ) -> None:
        dataset = dataset_w_features

        # In case of big number of iterations there are differences between CPU and GPU results
        base_model = AlternatingLeastSquares(factors=32, num_threads=2, use_gpu=use_gpu)
        self._init_model_factors_inplace(base_model, dataset)

>       model = ImplicitALSWrapperModel(model=base_model, fit_features_together=fit_features_together).fit(dataset)

tests/models/test_implicit_als.py:244:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
rectools/models/base.py:306: in fit
    self._fit(dataset, *args, **kwargs)
rectools/models/implicit_als.py:195: in _fit
    self._fit_model_for_epochs(dataset, self.model.iterations)
rectools/models/implicit_als.py:211: in _fit_model_for_epochs
    fit_als_with_features_together_inplace(
rectools/models/implicit_als.py:459: in fit_als_with_features_together_inplace
    _fit_combined_factors_on_gpu_inplace(
rectools/models/implicit_als.py:618: in _fit_combined_factors_on_gpu_inplace
    model.solver.calculate_yty(Y, _YtY, model.regularization)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

>   ???
E   ValueError: YtY and Y should have the same number of columns

_cuda.pyx:272: ValueError
=================================================== warnings summary ===================================================
tests/models/test_implicit_als.py: 20 warnings
  /home/aki/src/RecTools/rectools/dataset/identifiers.py:60: FutureWarning: unique with argument that is not not a Series, Index, ExtensionArray, or np.ndarray is deprecated and will raise in a future version.
    unq_values = pd.unique(values)

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
=============================================== short test summary info ================================================
FAILED tests/models/test_implicit_als.py::TestImplicitALSWrapperModel::test_happy_path_with_features[True-expected0-True] - ValueError: YtY and Y should have the same number of columns
====================================== 1 failed, 76 passed, 20 warnings in 11.30s ======================================

After

➜  RecTools git:(fix-als-gpu) poetry run pytest tests/models/test_implicit_als.py
================================================= test session starts ==================================================
platform linux -- Python 3.12.3, pytest-8.3.3, pluggy-1.5.0
rootdir: /home/aki/src/RecTools
configfile: setup.cfg
plugins: cov-5.0.0, subtests-0.12.1, typeguard-4.4.1, mock-3.14.0
collected 77 items

tests/models/test_implicit_als.py .............................................................................  [100%]

=================================================== warnings summary ===================================================
tests/models/test_implicit_als.py: 20 warnings
  /home/aki/src/RecTools/rectools/dataset/identifiers.py:60: FutureWarning: unique with argument that is not not a Series, Index, ExtensionArray, or np.ndarray is deprecated and will raise in a future version.
    unq_values = pd.unique(values)

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
=========================================== 77 passed, 20 warnings in 9.53s ============================================

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Optimization

How Has This Been Tested?

Before submitting a PR, please check yourself against the following list. It would save us quite a lot of time.

• Have you read the contribution guide?
• Have you updated the relevant docstrings? We're using Numpy format, please double-check yourself
• Does your change require any new tests?
• Have you updated the changelog file?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (9b3992e) to head (f9d317e).
Report is 99 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##              main      #228     +/-   ##
===========================================
  Coverage   100.00%   100.00%             
===========================================
  Files           45        59     +14     
  Lines         2242      3876   +1634     
===========================================
+ Hits          2242      3876   +1634     

☔ 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.

[blondered]

Missed this. Thank you!