def test_domain_range_scale_XYZ_to_Nayatani95(self):
"""
Test :func:`colour.appearance.nayatani95.XYZ_to_Nayatani95` definition
domain and range scale support.
"""
XYZ = np.array([19.01, 20.00, 21.78])
XYZ_n = np.array([95.05, 100.00, 108.88])
Y_o = 20.0
E_o = 5000.0
E_or = 1000.0
specification = XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or)
d_r = (
("reference", 1, 1),
("1", 0.01, np.array([1, 1, 1 / 360, 1, 1, 1, np.nan, np.nan, 1])),
(
"100",
1,
np.array([1, 1, 100 / 360, 1, 1, 1, np.nan, np.nan, 1]),
),
)
for scale, factor_a, factor_b in d_r:
with domain_range_scale(scale):
np.testing.assert_almost_equal(
XYZ_to_Nayatani95(XYZ * factor_a, XYZ_n * factor_a, Y_o,
E_o, E_or),
as_float_array(specification) * factor_b,
decimal=7,
)
def test_XYZ_to_Nayatani95(self):
"""
Test :func:`colour.appearance.nayatani95.XYZ_to_Nayatani95`
definition.
Notes
-----
- The test values have been generated from data of the following file
by *Fairchild (2013)*:
http://rit-mcsl.org/fairchild//files/AppModEx.xls
"""
XYZ = np.array([19.01, 20.00, 21.78])
XYZ_n = np.array([95.05, 100.00, 108.88])
Y_o = 20
E_o = 5000
E_or = 1000
np.testing.assert_allclose(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
np.array([50, 0.01, 257.5, 0.01, 62.6, 0.02, np.nan, np.nan, 50]),
rtol=0.01,
atol=0.01,
)
XYZ = np.array([57.06, 43.06, 31.96])
E_o = 500
np.testing.assert_allclose(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
np.array([73, 48.3, 21.6, 37.1, 67.3, 42.9, np.nan, np.nan, 75.9]),
rtol=0.01,
atol=0.01,
)
XYZ = np.array([3.53, 6.56, 2.14])
XYZ_n = np.array([109.85, 100.00, 35.58])
E_o = 5000
np.testing.assert_allclose(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
np.array(
[24.5, 49.3, 190.6, 81.3, 37.5, 62.1, np.nan, np.nan, 29.7]),
rtol=0.01,
atol=0.01,
)
XYZ = np.array([19.01, 20.00, 21.78])
E_o = 500
np.testing.assert_allclose(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
np.array(
[49.4, 39.9, 236.3, 40.2, 44.2, 35.8, np.nan, np.nan, 49.4]),
rtol=0.01,
atol=0.01,
)
def output_specification_from_data(self, data):
"""
Returns the *Nayatani (1995)* colour appearance model output
specification from given data.
Parameters
----------
data : list
Fixture data.
Returns
-------
Nayatani95_Specification
*Nayatani (1995)* colour appearance model specification.
"""
XYZ = tstack((data['X'], data['Y'], data['Z']))
XYZ_n = tstack((data['X_n'], data['Y_n'], data['Z_n']))
specification = XYZ_to_Nayatani95(XYZ,
XYZ_n,
data['Y_o'],
data['E_o'],
data['E_or'])
return specification
def test_n_dimensional_XYZ_to_Nayatani95(self):
"""
Test :func:`colour.appearance.nayatani95.XYZ_to_Nayatani95` definition
n-dimensional support.
"""
XYZ = np.array([19.01, 20.00, 21.78])
XYZ_n = np.array([95.05, 100.00, 108.88])
Y_o = 20
E_o = 5000
E_or = 1000
specification = XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or)
XYZ = np.tile(XYZ, (6, 1))
specification = np.tile(specification, (6, 1))
np.testing.assert_almost_equal(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
specification,
decimal=7,
)
XYZ_n = np.tile(XYZ_n, (6, 1))
np.testing.assert_almost_equal(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
specification,
decimal=7,
)
XYZ = np.reshape(XYZ, (2, 3, 3))
XYZ_n = np.reshape(XYZ_n, (2, 3, 3))
specification = np.reshape(specification, (2, 3, 9))
np.testing.assert_almost_equal(
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or),
specification,
decimal=7,
)
def test_nan_XYZ_to_Nayatani95(self):
"""
Test :func:`colour.appearance.nayatani95.XYZ_to_Nayatani95` definition
nan support.
"""
cases = [-1.0, 0.0, 1.0, -np.inf, np.inf, np.nan]
cases = set(permutations(cases * 3, r=3))
for case in cases:
XYZ = np.array(case)
XYZ_n = np.array(case)
Y_o = case[0]
E_o = case[0]
E_or = case[0]
XYZ_to_Nayatani95(XYZ, XYZ_n, Y_o, E_o, E_or)