Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

BUG: TestNanFunctions_Median.test_keepdims_out #22835

Closed
bnavigator opened this issue Dec 19, 2022 · 1 comment · Fixed by #22869
Closed

BUG: TestNanFunctions_Median.test_keepdims_out #22835

bnavigator opened this issue Dec 19, 2022 · 1 comment · Fixed by #22869
Labels
00 - Bug 05 - Testing
Milestone
1.24.1 release

Comments

@bnavigator
Copy link
Contributor

Describe the issue:

The random matrices for TestNanFunctions_Median.test_keepdims_out fail to slice ocassionally while building numpy 1.24 for openSUSE Tumbleweed.

Reproduce the code example:

import numpy as np
np.test(extra_argv=['-k', 'test_keepdims_out'])

Error message:

[ 2386s] _________________ TestNanFunctions_Median.test_keepdims_out[1] _________________
[ 2386s] [gw7] linux -- Python 3.8.16 /usr/bin/python3.8
[ 2386s] 
[ 2386s] self = <numpy.lib.tests.test_nanfunctions.TestNanFunctions_Median object at 0x7f07bc2c7520>
[ 2386s] axis = 1
[ 2386s] 
[ 2386s]     @pytest.mark.parametrize(
[ 2386s]         argnames='axis',
[ 2386s]         argvalues=[
[ 2386s]             None,
[ 2386s]             1,
[ 2386s]             (1, ),
[ 2386s]             (0, 1),
[ 2386s]             (-3, -1),
[ 2386s]         ]
[ 2386s]     )
[ 2386s]     def test_keepdims_out(self, axis):
[ 2386s]         d = np.ones((3, 5, 7, 11))
[ 2386s]         # Randomly set some elements to NaN:
[ 2386s]         w = np.random.random((4, 200)) * np.array(d.shape)[:, None]
[ 2386s]         w = w.astype(np.intp)
[ 2386s]         d[tuple(w)] = np.nan
[ 2386s]         if axis is None:
[ 2386s]             shape_out = (1,) * d.ndim
[ 2386s]         else:
[ 2386s]             axis_norm = normalize_axis_tuple(axis, d.ndim)
[ 2386s]             shape_out = tuple(
[ 2386s]                 1 if i in axis_norm else d.shape[i] for i in range(d.ndim))
[ 2386s]         out = np.empty(shape_out)
[ 2386s] >       result = np.nanmedian(d, axis=axis, keepdims=True, out=out)
[ 2386s] 
[ 2386s] axis       = 1
[ 2386s] axis_norm  = (1,)
[ 2386s] d          = array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ....1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1., nan,  1.]]]])
[ 2386s] out        = array([[[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-...73875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26]]]])
[ 2386s] self       = <numpy.lib.tests.test_nanfunctions.TestNanFunctions_Median object at 0x7f07bc2c7520>
[ 2386s] shape_out  = (3, 1, 7, 11)
[ 2386s] w          = array([[ 0,  1,  1,  1,  2,  1,  0,  0,  1,  1,  1,  1,  2,  2,  2,  0,
[ 2386s]          2,  1,  0,  0,  2,  0,  2,  1,  2,  0...,  9,
[ 2386s]          3,  4,  1, 10,  2, 10,  3,  6,  3,  7,  7, 10,  5,  3,  7,  0,
[ 2386s]          6,  0,  9,  9,  8,  6,  0,  1]])
[ 2386s] 
[ 2386s] ../../../BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/lib/tests/test_nanfunctions.py:834: 
[ 2386s] _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
[ 2386s] <__array_function__ internals>:200: in nanmedian
[ 2386s]     ???
[ 2386s]         args       = (array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ..., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1., nan,  1.]]]]),)
[ 2386s]         dispatcher = <function _nanmedian_dispatcher at 0x7f07c61a59d0>
[ 2386s]         implementation = <function nanmedian at 0x7f07c61a5af0>
[ 2386s]         kwargs     = {'axis': 1, 'keepdims': True, 'out': array([[[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.6367387...3875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26]]]])}
[ 2386s]         public_api = <function nanmedian at 0x7f07c61a5b80>
[ 2386s]         relevant_args = (array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ...3875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26]]]]))
[ 2386s] ../../../BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/lib/nanfunctions.py:1217: in nanmedian
[ 2386s]     return function_base._ureduce(a, func=_nanmedian, keepdims=keepdims,
[ 2386s]         a          = array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ....1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1., nan,  1.]]]])
[ 2386s]         axis       = 1
[ 2386s]         keepdims   = True
[ 2386s]         out        = array([[[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-...73875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26]]]])
[ 2386s]         overwrite_input = False
[ 2386s] ../../../BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/lib/function_base.py:3752: in _ureduce
[ 2386s]     r = func(a, **kwargs)
[ 2386s]         a          = array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ....1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1., nan,  1.]]]])
[ 2386s]         axis       = (1,)
[ 2386s]         func       = <function _nanmedian at 0x7f07c61a58b0>
[ 2386s]         index_out  = (slice(None, None, None), 0, slice(None, None, None), slice(None, None, None))
[ 2386s]         keepdims   = True
[ 2386s]         kwargs     = {'axis': 1, 'out': array([[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-...3673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26]]]), 'overwrite_input': False}
[ 2386s]         nd         = 4
[ 2386s]         out        = array([[[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-...73875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]           -8.63673875e-26, -8.63673875e-26]]]])
[ 2386s] ../../../BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/lib/nanfunctions.py:1094: in _nanmedian
[ 2386s]     return _nanmedian_small(a, axis, out, overwrite_input)
[ 2386s]         a          = array([[[[ 1., nan,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ..., nan,  1.,  1.],
[ 2386s]          [nan,  1.,  1., ....1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1.,  1.,  1.],
[ 2386s]          [ 1.,  1.,  1., ...,  1., nan,  1.]]]])
[ 2386s]         axis       = 1
[ 2386s]         out        = array([[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26...63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26]]])
[ 2386s]         overwrite_input = False
[ 2386s] _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
[ 2386s] 
[ 2386s] a = masked_array(
[ 2386s]   data=[[[[1.0, --, 1.0, ..., 1.0, 1.0, 1.0],
[ 2386s]           [1.0, 1.0, 1.0, ..., --, 1.0, 1.0],
[ 2386s]           [-...se, False, ..., False, False, False],
[ 2386s]           [False, False, False, ..., False,  True, False]]]],
[ 2386s]   fill_value=1e+20)
[ 2386s] axis = 1
[ 2386s] out = array([[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26...63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26]]])
[ 2386s] overwrite_input = False
[ 2386s] 
[ 2386s]     def _nanmedian_small(a, axis=None, out=None, overwrite_input=False):
[ 2386s]         """
[ 2386s]         sort + indexing median, faster for small medians along multiple
[ 2386s]         dimensions due to the high overhead of apply_along_axis
[ 2386s]     
[ 2386s]         see nanmedian for parameter usage
[ 2386s]         """
[ 2386s]         a = np.ma.masked_array(a, np.isnan(a))
[ 2386s]         m = np.ma.median(a, axis=axis, overwrite_input=overwrite_input)
[ 2386s]         for i in range(np.count_nonzero(m.mask.ravel())):
[ 2386s] >           warnings.warn("All-NaN slice encountered", RuntimeWarning,
[ 2386s]                           stacklevel=4)
[ 2386s] E           RuntimeWarning: All-NaN slice encountered
[ 2386s] 
[ 2386s] a          = masked_array(
[ 2386s]   data=[[[[1.0, --, 1.0, ..., 1.0, 1.0, 1.0],
[ 2386s]           [1.0, 1.0, 1.0, ..., --, 1.0, 1.0],
[ 2386s]           [-...se, False, ..., False, False, False],
[ 2386s]           [False, False, False, ..., False,  True, False]]]],
[ 2386s]   fill_value=1e+20)
[ 2386s] axis       = 1
[ 2386s] i          = 0
[ 2386s] m          = masked_array(
[ 2386s]   data=[[[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
[ 2386s]          [1.0, 1.0, 1.0, 1.0, 1.0, 1.0...         [False, False, False, False, False, False, False, False, False,
[ 2386s]           False, False]]],
[ 2386s]   fill_value=1e+20)
[ 2386s] out        = array([[[-8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26...63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26, -8.63673875e-26,
[ 2386s]          -8.63673875e-26, -8.63673875e-26]]])
[ 2386s] overwrite_input = False
[ 2386s] 
[ 2386s] ../../../BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/lib/nanfunctions.py/home/abuild/rpmbuild/BUILDROOT/python-numpy-1.24.0-124.1.x86_64/usr/lib64/python3.8/site-packages/numpy/_pytesttester.py:143: DeprecationWarning: 
[ 2386s] 
[ 2386s]   `numpy.distutils` is deprecated since NumPy 1.23.0, as a result
[ 2386s]   of the deprecation of `distutils` itself. It will be removed for
[ 2386s]   Python >= 3.12. For older Python versions it will remain present.
[ 2386s]   It is recommended to use `setuptools < 60.0` for those Python versions.
[ 2386s]   For more details, see:
[ 2386s]     https://numpy.org/devdocs/reference/distutils_status_migration.html 
[ 2386s] 
[ 2386s] 
[ 2386s]   from numpy.distutils import cpuinfo
[ 2386s] :1111: RuntimeWarning
[ 2386s] =========================== short test summary info ============================
[ 2386s] FAILED lib/tests/test_nanfunctions.py::TestNanFunctions_Median::test_keepdims_out[1]
[ 2386s] = 1 failed, 23068 passed, 2238 skipped, 35 xfailed, 3 xpassed in 436.92s (0:07:16) =

NumPy/Python version information:

1.24.0 3.10.9 (main, Dec 08 2022, 14:49:06) [GCC]

Context for the issue:

This is a flaky issue. Sometimes the test passes, sometimes not. On different parametrizations.
@seberg
Copy link
Member

seberg commented Dec 19, 2022

This failure is harmless, but the warning should be ignored (or a seed used to ensure its not random).

@seberg seberg added this to the 1.24.1 release milestone Dec 20, 2022
seberg added a commit to seberg/numpy that referenced this issue Dec 22, 2022
@seberg
TST: Ignore nan-warnings in randomized out tests
48623f6 
The tests randomize the nan pattern and thus can run into these
(additional) warnings, so ignore them.
(Could also fix the random seed, but this should do)

Closes numpygh-22835
@seberg seberg mentioned this issue Dec 22, 2022
Merged
charris pushed a commit to charris/numpy that referenced this issue Dec 22, 2022
@seberg @charris
TST: Ignore nan-warnings in randomized out tests
8e2840a 
The tests randomize the nan pattern and thus can run into these
(additional) warnings, so ignore them.
(Could also fix the random seed, but this should do)

Closes numpygh-22835
charris pushed a commit to charris/numpy that referenced this issue Dec 22, 2022
@seberg @charris
TST: Ignore nan-warnings in randomized out tests
dc7bac6 
The tests randomize the nan pattern and thus can run into these
(additional) warnings, so ignore them.
(Could also fix the random seed, but this should do)

Closes numpygh-22835
@charris charris mentioned this issue Dec 22, 2022
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
00 - Bug 05 - Testing
Projects
None yet
Milestone
1.24.1 release
Development

Successfully merging a pull request may close this issue.

TST: Ignore nan-warnings in randomized nanfunction out= tests seberg/numpy
2 participants
@seberg @bnavigator