search

AnacondaRecipes / scikit-learn-feedstock

A conda-smithy repository for scikit-learn.
BSD 3-Clause "New" or "Revised" License
0 stars 2 forks source link

scikit-learn 0.24.1: broken tests on win-32 #3

Closed chenghlee closed 2 years ago

chenghlee commented 3 years ago 
=========================== short test summary info ===========================
FAILED cluster/tests/test_spectral.py::test_precomputed_nearest_neighbors_filtering
FAILED manifold/tests/test_spectral_embedding.py::test_precomputed_nearest_neighbors_filtering
FAILED tests/test_common.py::test_estimators[GaussianProcessRegressor()-check_supervised_y_2d]
FAILED tests/test_common.py::test_estimators[GaussianProcessRegressor()-check_fit_idempotent]
FAILED tests/test_common.py::test_estimators[Nystroem()-check_fit_idempotent]
FAILED tests/test_common.py::test_estimators[SpectralEmbedding()-check_pipeline_consistency]
= 6 failed, 17941 passed, 996 skipped, 28 xfailed, 24 xpassed, 3047 warnings in 1083.37s (0:18:03) =

================================== FAILURES ===================================
________________ test_precomputed_nearest_neighbors_filtering _________________
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

    def test_precomputed_nearest_neighbors_filtering():
        ### ... snip ... ###
>       assert_array_equal(results[0], results[1])
E       AssertionError: 
E       Arrays are not equal
E       
E       Mismatched elements: 47 / 200 (23.5%)
E       Max absolute difference: 1
E       Max relative difference: 1.
E        x: array([1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
E              1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1,
E              1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0,...
E        y: array([0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 1,
E              1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1,
E              1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0,...

..\_test_env\lib\site-packages\sklearn\cluster\tests\test_spectral.py:123: AssertionError

________________ test_precomputed_nearest_neighbors_filtering _________________
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

    def test_precomputed_nearest_neighbors_filtering():
        ### ...snip... ###
>       assert_array_equal(results[0], results[1])
E       AssertionError: 
E       Arrays are not equal
E       
E       Mismatched elements: 2000 / 2000 (100%)
E       Max absolute difference: 0.202388
E       Max relative difference: 107.51598505
E        x: array([[-0.006769, -0.007541],
E              [-0.01462 , -0.037366],
E              [-0.025658, -0.009109],...
E        y: array([[ 0.011987, -0.030262],
E              [-0.010395,  0.032586],
E              [ 0.036888, -0.033157],...

..\_test_env\lib\site-packages\sklearn\manifold\tests\test_spectral_embedding.py:155: AssertionError

______ test_estimators[GaussianProcessRegressor()-check_supervised_y_2d] ______
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

estimator = GaussianProcessRegressor()
check = functools.partial(<function check_supervised_y_2d at 0x06932340>, 'GaussianProcessRegressor')
request = <FixtureRequest for <Function test_estimators[GaussianProcessRegressor()-check_supervised_y_2d]>>

    @parametrize_with_checks(list(_tested_estimators()))
    def test_estimators(estimator, check, request):
        # Common tests for estimator instances
        with ignore_warnings(category=(FutureWarning,
                                       ConvergenceWarning,
                                       UserWarning, FutureWarning)):
            _set_checking_parameters(estimator)
>           check(estimator)

..\_test_env\lib\site-packages\sklearn\tests\test_common.py:93: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..\_test_env\lib\site-packages\sklearn\utils\_testing.py:308: in wrapper
    return fn(*args, **kwargs)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

name = 'GaussianProcessRegressor', estimator_orig = GaussianProcessRegressor()

    @ignore_warnings(category=FutureWarning)
    def check_supervised_y_2d(name, estimator_orig):
        ### ...snip... ###
>       assert_allclose(y_pred.ravel(), y_pred_2d.ravel())
E       AssertionError: 
E       Not equal to tolerance rtol=1e-07, atol=0
E       
E       Mismatched elements: 10 / 30 (33.3%)
E       Max absolute difference: 3.51023877e-10
E       Max relative difference: 0.00033019
E        x: array([-1.455964e-05,  1.000040e+00,  2.000001e+00,  1.030063e-05,
E               1.000001e+00,  2.000004e+00,  1.870067e-07,  9.999929e-01,
E               2.000001e+00, -2.525699e-06,  1.000002e+00,  2.000025e+00,...
E        y: array([-1.455970e-05,  1.000040e+00,  2.000001e+00,  1.030057e-05,
E               1.000001e+00,  2.000004e+00,  1.869463e-07,  9.999929e-01,
E               2.000001e+00, -2.525771e-06,  1.000002e+00,  2.000025e+00,...

..\_test_env\lib\site-packages\sklearn\utils\estimator_checks.py:2184: AssertionError

______ test_estimators[GaussianProcessRegressor()-check_fit_idempotent] _______
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

estimator = GaussianProcessRegressor()
check = functools.partial(<function check_fit_idempotent at 0x0692A070>, 'GaussianProcessRegressor')
request = <FixtureRequest for <Function test_estimators[GaussianProcessRegressor()-check_fit_idempotent]>>

    @parametrize_with_checks(list(_tested_estimators()))
    def test_estimators(estimator, check, request):
        # Common tests for estimator instances
        with ignore_warnings(category=(FutureWarning,
                                       ConvergenceWarning,
                                       UserWarning, FutureWarning)):
            _set_checking_parameters(estimator)
>           check(estimator)

..\_test_env\lib\site-packages\sklearn\tests\test_common.py:93: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..\_test_env\lib\site-packages\sklearn\utils\estimator_checks.py:3031: in check_fit_idempotent
    assert_allclose_dense_sparse(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

x = array([-5.99002466e+00, -1.10709372e+00,  9.46448649e-01, -2.46598440e+00,
        3.73609516e+00, -4.17400605e+01, -7...5208e+00, -4.58927614e+02,  3.76215227e+02,
        4.83375571e-01,  7.75619447e+00, -4.39485206e+00, -2.42046605e+00])
y = array([-5.99002008e+00, -1.10709393e+00,  9.46448479e-01, -2.46598012e+00,
        3.73608357e+00, -4.17399150e+01, -7...4839e+00, -4.58927694e+02,  3.76214846e+02,
        4.83375105e-01,  7.75618923e+00, -4.39485343e+00, -2.42046461e+00])
rtol = 1e-07, atol = 1e-09
err_msg = 'Idempotency check failed for method predict'

    def assert_allclose_dense_sparse(x, y, rtol=1e-07, atol=1e-9, err_msg=''):
        ### ...snip... ###
>           assert_allclose(x, y, rtol=rtol, atol=atol, err_msg=err_msg)
E           AssertionError: 
E           Not equal to tolerance rtol=1e-07, atol=1e-09
E           Idempotency check failed for method predict
E           Mismatched elements: 20 / 20 (100%)
E           Max absolute difference: 0.00072748
E           Max relative difference: 3.48470055e-06
E            x: array([-5.990025e+00, -1.107094e+00,  9.464486e-01, -2.465984e+00,
E                   3.736095e+00, -4.174006e+01, -7.626134e+02,  2.480707e+00,
E                  -1.449927e+03, -1.408592e+00,  2.477522e+00, -6.230015e+00,...
E            y: array([-5.990020e+00, -1.107094e+00,  9.464485e-01, -2.465980e+00,
E                   3.736084e+00, -4.173992e+01, -7.626135e+02,  2.480703e+00,
E                  -1.449926e+03, -1.408593e+00,  2.477524e+00, -6.230021e+00,...

..\_test_env\lib\site-packages\sklearn\utils\_testing.py:415: AssertionError

______________ test_estimators[Nystroem()-check_fit_idempotent] _______________
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

estimator = Nystroem()
check = functools.partial(<function check_fit_idempotent at 0x0692A070>, 'Nystroem')
request = <FixtureRequest for <Function test_estimators[Nystroem()-check_fit_idempotent]>>

    @parametrize_with_checks(list(_tested_estimators()))
    def test_estimators(estimator, check, request):
        # Common tests for estimator instances
        with ignore_warnings(category=(FutureWarning,
                                       ConvergenceWarning,
                                       UserWarning, FutureWarning)):
            _set_checking_parameters(estimator)
>           check(estimator)

..\_test_env\lib\site-packages\sklearn\tests\test_common.py:93: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..\_test_env\lib\site-packages\sklearn\utils\estimator_checks.py:3031: in check_fit_idempotent
    assert_allclose_dense_sparse(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

x = array([[ 1.97641385e-01,  9.61580107e-05,  3.36263817e-01, ...,
         2.12565982e-01,  3.22560841e-02,  1.32599218e...  [ 6.55313722e-02, -3.43879184e-04,  2.22385697e-02, ...,
         6.38597798e-03,  5.60496364e-01,  1.01087809e-02]])
y = array([[ 1.97641385e-01,  9.61580106e-05,  3.36263817e-01, ...,
         2.12565982e-01,  3.22560841e-02,  1.32599218e...  [ 6.55313724e-02, -3.43879238e-04,  2.22385701e-02, ...,
         6.38597857e-03,  5.60496364e-01,  1.01087809e-02]])
rtol = 1e-07, atol = 1e-09
err_msg = 'Idempotency check failed for method transform'

    def assert_allclose_dense_sparse(x, y, rtol=1e-07, atol=1e-9, err_msg=''):
        ### ...snip... ###
>           assert_allclose(x, y, rtol=rtol, atol=atol, err_msg=err_msg)
E           AssertionError: 
E           Not equal to tolerance rtol=1e-07, atol=1e-09
E           Idempotency check failed for method transform
E           Mismatched elements: 338 / 1600 (21.1%)
E           Max absolute difference: 2.3087952e-06
E           Max relative difference: 0.00080639
E            x: array([[ 1.976414e-01,  9.615801e-05,  3.362638e-01, ...,  2.125660e-01,
E                    3.225608e-02,  1.325992e-01],
E                  [ 3.464009e-03,  5.946941e-03,  1.600314e-03, ...,  7.684868e-04,...
E            y: array([[ 1.976414e-01,  9.615801e-05,  3.362638e-01, ...,  2.125660e-01,
E                    3.225608e-02,  1.325992e-01],
E                  [ 3.464004e-03,  5.946940e-03,  1.600314e-03, ...,  7.684816e-04,...

..\_test_env\lib\site-packages\sklearn\utils\_testing.py:415: AssertionError

_______ test_estimators[SpectralEmbedding()-check_pipeline_consistency] _______
[gw0] win32 -- Python 3.9.2 %PREFIX%\python.exe

estimator = SpectralEmbedding()
check = functools.partial(<function check_pipeline_consistency at 0x06925898>, 'SpectralEmbedding')
request = <FixtureRequest for <Function test_estimators[SpectralEmbedding()-check_pipeline_consistency]>>

    @parametrize_with_checks(list(_tested_estimators()))
    def test_estimators(estimator, check, request):
        # Common tests for estimator instances
        with ignore_warnings(category=(FutureWarning,
                                       ConvergenceWarning,
                                       UserWarning, FutureWarning)):
            _set_checking_parameters(estimator)
>           check(estimator)

..\_test_env\lib\site-packages\sklearn\tests\test_common.py:93: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..\_test_env\lib\site-packages\sklearn\utils\_testing.py:308: in wrapper
    return fn(*args, **kwargs)
..\_test_env\lib\site-packages\sklearn\utils\estimator_checks.py:1406: in check_pipeline_consistency
    assert_allclose_dense_sparse(result, result_pipe)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

x = array([[ 0.08333619,  0.16885926],
       [ 0.16642444, -0.07159973],
       [ 0.16642444, -0.07159973],
       [-0.02...-0.07159973],
       [ 0.08333619,  0.16885926],
       [ 0.08333619,  0.16885926],
       [ 0.16642444, -0.07159973]])
y = array([[ 9.53543159e-02,  3.00838180e-15],
       [ 1.60809881e-01, -9.43588019e-02],
       [ 1.60809881e-01,  2.6359....53543159e-02,  2.21466217e-16],
       [ 9.53543159e-02, -6.94851237e-16],
       [ 1.60809881e-01, -2.12784505e-01]])
rtol = 1e-07, atol = 1e-09, err_msg = ''

    def assert_allclose_dense_sparse(x, y, rtol=1e-07, atol=1e-9, err_msg=''):
        ### ...snip... ###
>           assert_allclose(x, y, rtol=rtol, atol=atol, err_msg=err_msg)
E           AssertionError: 
E           Not equal to tolerance rtol=1e-07, atol=1e-09
E           
E           Mismatched elements: 60 / 60 (100%)
E           Max absolute difference: 0.35748972
E           Max relative difference: 1.75526823e+16
E            x: array([[ 0.083336,  0.168859],
E                  [ 0.166424, -0.0716  ],
E                  [ 0.166424, -0.0716  ],...
E            y: array([[ 9.535432e-02,  3.008382e-15],
E                  [ 1.608099e-01, -9.435880e-02],
E                  [ 1.608099e-01,  2.635916e-01],...
..\_test_env\lib\site-packages\sklearn\utils\_testing.py:415: AssertionError
cbouss commented 2 years ago

win-32 support has been dropped.