def test_unity_3x3(self, boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, fft_type):
'''
Test that a 3x3 unit kernel returns the same array (except when
boundary is None).
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([[1., 2., 3.],
[4., 5., 6.],
[7., 8., 9.]], dtype='float64')
y = np.array([[0., 0., 0.],
[0., 1., 0.],
[0., 0., 0.]], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
assert_array_almost_equal_nulp(z, x, 10)
def generate_or_test(self, generate, figure, image, adjust_bbox=True):
if generate is None:
result_dir = tempfile.mkdtemp()
test_image = os.path.abspath(os.path.join(result_dir, image))
# distutils will put the baseline images in non-accessible places,
# copy to our tmpdir to be sure to keep them in case of failure
orig_baseline_image = os.path.abspath(os.path.join(self._baseline_images_dir, image))
baseline_image = os.path.abspath(os.path.join(result_dir, 'baseline-'+image))
shutil.copyfile(orig_baseline_image, baseline_image)
figure.save(test_image)
if not os.path.exists(baseline_image):
raise Exception("""Image file not found for comparision test
Generated Image:
\t{test}
This is expected for new tests.""".format(
test=test_image))
msg = compare_images(baseline_image, test_image, tol=self._tolerance)
if msg is None:
shutil.rmtree(result_dir)
else:
raise Exception(msg)
else:
figure.save(os.path.abspath(os.path.join(generate, image)), adjust_bbox=adjust_bbox)
pytest.skip("Skipping test, since generating data")
def generate_or_test(self, generate, figure, image, adjust_bbox=True):
if generate is None:
result_dir = tempfile.mkdtemp()
test_image = os.path.abspath(os.path.join(result_dir, image))
# distutils will put the baseline images in non-accessible places,
# copy to our tmpdir to be sure to keep them in case of failure
orig_baseline_image = os.path.abspath(
os.path.join(self._baseline_images_dir, image))
baseline_image = os.path.abspath(
os.path.join(result_dir, 'baseline-' + image))
shutil.copyfile(orig_baseline_image, baseline_image)
figure.save(test_image)
if not os.path.exists(baseline_image):
raise Exception("""Image file not found for comparision test
Generated Image:
\t{test}
This is expected for new tests.""".format(
test=test_image))
msg = compare_images(baseline_image,
test_image,
tol=self._tolerance)
if msg is None:
shutil.rmtree(result_dir)
else:
raise Exception(msg)
else:
figure.save(os.path.abspath(os.path.join(generate, image)),
adjust_bbox=adjust_bbox)
pytest.skip("Skipping test, since generating data")
def test_unity_1_withnan(self, boundary, interpolate_nan, normalize_kernel,
ignore_edge_zeros, fft_type):
'''
Test that a unit kernel with three elements returns the same array
(except when boundary is None). This version includes a NaN value in
the original array.
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([1., np.nan, 3.], dtype='float64')
y = np.array([1.], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
if fft_type == 'numpy':
# for whatever reason, numpy's fft has very limited precision, and
# the comparison fails unless you cast the float64 to a float16
# np1.4 incompatible assert_array_almost_equal_nulp(np.asarray(z, dtype=np.float16), np.array([1.,0.,3.], dtype=np.float16), 10)
assert np.all(np.abs(z - np.array([1., 0., 3.])) < 1e-14)
else:
assert_array_almost_equal_nulp(z, np.array([1., 0., 3.], dtype='float64'), 10)
def test_uniform_3x3(self, boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, fft_type):
'''
Test that the different modes are producing the correct results using
a 3x3 uniform kernel.
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([[0., 0., 3.],
[1., 0., 0.],
[0., 2., 0.]], dtype='float64')
y = np.array([[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.]], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
w = np.array([[4., 6., 4.],
[6., 9., 6.],
[4., 6., 4.]], dtype='float64')
answer_dict = {
'sum': np.array([[1., 4., 3.],
[3., 6., 5.],
[3., 3., 2.]], dtype='float64'),
'sum_wrap': np.array([[6., 6., 6.],
[6., 6., 6.],
[6., 6., 6.]], dtype='float64'),
}
answer_dict['average'] = answer_dict['sum'] / w
answer_dict['average_wrap'] = answer_dict['sum_wrap'] / 9.
answer_dict['average_withzeros'] = answer_dict['sum'] / 9.
answer_dict['sum_withzeros'] = answer_dict['sum']
if normalize_kernel:
answer_key = 'average'
else:
answer_key = 'sum'
if boundary == 'wrap':
answer_key += '_wrap'
elif not ignore_edge_zeros:
answer_key += '_withzeros'
a = answer_dict[answer_key]
print boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, answer_key
print "z: ", z
print "answer: ", a
print "ratio: ", z / a
# for reasons unknown, the Windows FFT returns an answer for the [0,0]
# component that is EXACTLY 10*np.spacing
assert np.all(np.abs(z - a) <= np.spacing(np.where(z > a, z, a)) * 10)
def test_uniform_3_withnan(self, boundary, interpolate_nan,
normalize_kernel, ignore_edge_zeros, fft_type):
'''
Test that the different modes are producing the correct results using
a uniform kernel with three elements. This version includes a NaN
value in the original array.
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([1., np.nan, 3.], dtype='float64')
y = np.array([1., 1., 1.], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
answer_dict = {
'sum': np.array([1., 4., 3.], dtype='float64'),
'sum_nozeros': np.array([1., 4., 3.], dtype='float64'),
'sum_zeros': np.array([1., 4., 3.], dtype='float64'),
'sum_zeros_noignan': np.array([1., 4., 3.], dtype='float64'),
'sum_nozeros_noignan': np.array([1., 4., 3.], dtype='float64'),
'average': np.array([1., 2., 3.], dtype='float64'),
'sum_wrap': np.array([4., 4., 4.], dtype='float64'),
'sum_wrap_noignan': np.array([4., 4., 4.], dtype='float64'),
'average_wrap': np.array([(1 + 3) / 2., 2., 2.], dtype='float64'),
'average_wrap_noignan': np.array([4 / 3., 4 / 3., 4 / 3.], dtype='float64'),
'average_nozeros': np.array([1, 2, 3], dtype='float64'),
'average_nozeros_noignan': np.array([1 / 2., 4 / 3., 3 / 2.], dtype='float64'),
'average_zeros': np.array([1 / 2., 4 / 2., 3 /2.], dtype='float64'),
'average_zeros_noignan': np.array([1/3., 4/3., 3/3.], dtype='float64'),
}
if normalize_kernel:
answer_key = 'average'
else:
answer_key = 'sum'
if boundary == 'wrap':
answer_key += '_wrap'
elif ignore_edge_zeros:
answer_key += '_nozeros'
else:
# average = average_zeros; sum = sum_zeros
answer_key += '_zeros'
if not interpolate_nan:
answer_key += '_noignan'
print boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, answer_key
assert_array_almost_equal_nulp(z, answer_dict[answer_key], 10)
def test_helper_poisson_exact():
pytest.skip('distribution stretch goal not yet implemented')
centerq = [1, 5, 30, 400] * u.one
ds.poisson(centerq, n_samples=1000)
with pytest.raises(u.UnitsError) as exc:
centerq = [1, 5, 30, 400] * u.kpc
ds.poisson(centerq, n_samples=1000)
assert exc.value.args[0] == ("Poisson distribution can only be computed "
"for dimensionless quantities")
def pytest_runtest_setup(item):
if 'generate_reference' in item.keywords:
if item.config.getvalue("generate_reference"):
if item.config.getvalue("generate_reference").startswith('-'):
raise Exception("Need to specify output directory for generating reference files")
if not item.config.getvalue("generate_reference").startswith('/'):
raise Exception("Need to specify output directory for generating reference files as an absolute path")
item.funcargs['generate'] = item.config.getvalue("generate_reference")
if 'enable_bit_level_tests' in item.keywords and not item.config.getvalue("enable_bit_level_tests"):
pytest.skip("need --enable-bit-level-tests option to run")
def test_helper_poisson_exact():
pytest.skip('distribution stretch goal not yet implemented')
centerq = [1, 5, 30, 400] * u.one
ds.poisson(centerq, n_samples=1000)
with pytest.raises(u.UnitsError) as exc:
centerq = [1, 5, 30, 400] * u.kpc
ds.poisson(centerq, n_samples=1000)
assert exc.value.args[0] == ("Poisson distribution can only be computed "
"for dimensionless quantities")
def test_array_repr_latex():
# as of this writing ndarray does not have a _repr_latex_, and this test
# ensure distributions account for that. However, if in the future ndarray
# gets a _repr_latex_, we can skip this.
arr = np.random.randn(4, 1000)
if hasattr(arr, '_repr_latex_'):
pytest.skip('in this version of numpy, ndarray has a _repr_latex_')
distr = Distribution(arr)
assert distr._repr_latex_() is None
def test_mask_area(self):
try:
mask = self.reg.to_mask(mode='exact')
assert_allclose(np.sum(mask.data), self.expected_area)
except NotImplementedError:
try:
mask = self.reg.to_mask(mode='subpixels', subpixels=30)
assert_allclose(np.sum(mask.data),
self.expected_area,
rtol=0.005)
except NotImplementedError:
pytest.skip()
def test_array_repr_latex():
# as of this writing ndarray does not have a _repr_latex_, and this test
# ensure distributions account for that. However, if in the future ndarray
# gets a _repr_latex_, we can skip this.
arr = np.random.randn(4, 1000)
if hasattr(arr, '_repr_latex_'):
pytest.skip('in this version of numpy, ndarray has a _repr_latex_')
distr = Distribution(arr)
assert distr._repr_latex_() is None
def pytest_runtest_setup(item):
if 'generate_reference' in item.keywords:
if item.config.getvalue("generate_reference"):
if item.config.getvalue("generate_reference").startswith('-'):
raise Exception(
"Need to specify output directory for generating reference files"
)
if not item.config.getvalue("generate_reference").startswith('/'):
raise Exception(
"Need to specify output directory for generating reference files as an absolute path"
)
item.funcargs['generate'] = item.config.getvalue(
"generate_reference")
if 'enable_bit_level_tests' in item.keywords and not item.config.getvalue(
"enable_bit_level_tests"):
pytest.skip("need --enable-bit-level-tests option to run")
def test_unity_1_none(self, boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, fft_type):
'''
Test that a unit kernel with a single element returns the same array
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([1., 2., 3.], dtype='float64')
y = np.array([1.], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
assert_array_almost_equal_nulp(z, x, 10)
def test_uniform_3x3_withnan(self, boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, fft_type):
'''
Test that the different modes are producing the correct results using
a 3x3 uniform kernel. This version includes a NaN value in the
original array.
'''
if fft_type == 'fftw' and not HAS_FFTW:
pytest.skip('fftw3 is not installed')
elif fft_type == 'scipy' and not HAS_SCIPY:
pytest.skip('scipy is not installed')
x = np.array([[0., 0., 3.],
[1., np.nan, 0.],
[0., 2., 0.]], dtype='float64')
y = np.array([[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.]], dtype='float64')
z = convolve_fft(x, y, boundary=boundary,
interpolate_nan=interpolate_nan,
normalize_kernel=normalize_kernel,
ignore_edge_zeros=ignore_edge_zeros,
fft_type=fft_type)
w_n = np.array([[3., 5., 3.],
[5., 8., 5.],
[3., 5., 3.]], dtype='float64')
w_z = np.array([[4., 6., 4.],
[6., 9., 6.],
[4., 6., 4.]], dtype='float64')
answer_dict = {
'sum': np.array([[1., 4., 3.],
[3., 6., 5.],
[3., 3., 2.]], dtype='float64'),
'sum_wrap': np.array([[6., 6., 6.],
[6., 6., 6.],
[6., 6., 6.]], dtype='float64'),
}
answer_dict['average'] = answer_dict['sum'] / w_z
answer_dict['average_ignan'] = answer_dict['sum'] / w_n
answer_dict['average_wrap_ignan'] = answer_dict['sum_wrap'] / 8.
answer_dict['average_wrap'] = answer_dict['sum_wrap'] / 9.
answer_dict['average_withzeros'] = answer_dict['sum'] / 9.
answer_dict['average_withzeros_ignan'] = answer_dict['sum'] / 8.
answer_dict['sum_withzeros'] = answer_dict['sum']
answer_dict['sum_ignan'] = answer_dict['sum']
answer_dict['sum_withzeros_ignan'] = answer_dict['sum']
answer_dict['sum_wrap_ignan'] = answer_dict['sum_wrap']
if normalize_kernel:
answer_key = 'average'
else:
answer_key = 'sum'
if boundary == 'wrap':
answer_key += '_wrap'
elif not ignore_edge_zeros:
answer_key += '_withzeros'
if interpolate_nan:
answer_key += '_ignan'
a = answer_dict[answer_key]
print boundary, interpolate_nan, normalize_kernel, ignore_edge_zeros, answer_key
print "z: ", z
print "answer: ", a
print "ratio: ", z / a
assert np.all(np.abs(z - a) < np.spacing(np.where(z > a, z, a)) * 10)
def test_helper_normal_exact():
pytest.skip('distribution stretch goal not yet implemented')
centerq = [1, 5, 30, 400] * u.kpc
ds.normal(centerq, std=[0.2, 1.5, 4, 1]*u.kpc)
ds.normal(centerq, var=[0.04, 2.25, 16, 1]*u.kpc**2)
ds.normal(centerq, ivar=[25, 0.44444444, 0.625, 1]*u.kpc**-2)
def test_area(self):
try:
assert_allclose(self.reg.area, self.expected_area)
except ImportError:
pytest.skip()
def test_helper_normal_exact():
pytest.skip('distribution stretch goal not yet implemented')
centerq = [1, 5, 30, 400] * u.kpc
ds.normal(centerq, std=[0.2, 1.5, 4, 1]*u.kpc)
ds.normal(centerq, var=[0.04, 2.25, 16, 1]*u.kpc**2)
ds.normal(centerq, ivar=[25, 0.44444444, 0.625, 1]*u.kpc**-2)
def test_pix_area(region):
try:
area = region.area
assert not isinstance(area, u.Quantity)
except ImportError: # for shapely
pytest.skip()
