def test_generalizedOM(self):
# generalized OM model
gom = Anisotropy(anisotropy_type='GOM')
r = self._r_array
R = 2
om = Anisotropy(anisotropy_type='OM')
kwargs_om = {'r_ani': 1.}
kwargs_gom = {'r_ani': 1., 'beta_inf': 1.}
beta_gom = gom.beta_r(r, **kwargs_gom)
beta_om = om.beta_r(r, **kwargs_om)
npt.assert_almost_equal(beta_gom, beta_om, decimal=5)
f_om = om.anisotropy_solution(r, **kwargs_om)
f_gom = gom.anisotropy_solution(r, **kwargs_gom)
npt.assert_almost_equal(f_gom, f_om, decimal=5)
K_gom = gom.K(r, R, **kwargs_gom)
K_om = om.K(r, R, **kwargs_om)
npt.assert_almost_equal(K_gom, K_om, decimal=3)
assert hasattr(gom._model, '_f_12_interp')
assert hasattr(gom._model, '_f_32_interp')
gom.delete_anisotropy_cache()
if hasattr(gom._model, '_f_12_interp'):
assert False
if hasattr(gom._model, '_f_32_interp'):
assert False
from lenstronomy.GalKin.anisotropy import GeneralizedOM
gom_class = GeneralizedOM()
_F = gom_class._F(a=3 / 2., z=0.5, beta_inf=1)
_F_array = gom_class._F(a=3 / 2., z=np.array([0.5]), beta_inf=1)
npt.assert_almost_equal(_F_array[0], _F, decimal=5)
def test_K(self):
anisoClass = Anisotropy(anisotropy_type='const')
kwargs = {'beta': 1.0}
k = anisoClass.K(self._r_array, self._R_array, **kwargs)
npt.assert_almost_equal(k[0], 0.61418484930437822, decimal=5)
anisoClass = Anisotropy(anisotropy_type='Colin')
kwargs = {'r_ani': 1}
k = anisoClass.K(self._r_array, self._R_array, **kwargs)
npt.assert_almost_equal(k[0], 0.91696135187291117, decimal=5)
k = anisoClass.K(self._r_array, self._R_array - 0.001, **kwargs)
npt.assert_almost_equal(k[0], 0.91696135187291117, decimal=2)
k = anisoClass.K(self._r_array, self._R_array + 0.001, **kwargs)
npt.assert_almost_equal(k[0], 0.91696135187291117, decimal=2)
anisoClass = Anisotropy(anisotropy_type='radial')
kwargs = {}
k = anisoClass.K(self._r_array, self._R_array, **kwargs)
npt.assert_almost_equal(k[0], 0.61418484930437856, decimal=5)
anisoClass = Anisotropy(anisotropy_type='isotropic')
kwargs = {}
k = anisoClass.K(self._r_array, self._R_array, **kwargs)
npt.assert_almost_equal(k[0], 0.8660254037844386, decimal=5)
anisoClass = Anisotropy(anisotropy_type='OM')
kwargs = {'r_ani': 1}
k = anisoClass.K(self._r_array, self._R_array, **kwargs)
npt.assert_almost_equal(k[0], 0.95827704196894481, decimal=5)
R = 0.1
k = anisoClass.K(R, R, **kwargs)
npt.assert_almost_equal(k, 0, decimal=5)
def test_isotropic_anisotropy(self):
# radial
r = 2.
R = 1.
isotropic = Anisotropy(anisotropy_type='isotropic')
kwargs_iso = {}
beta = isotropic.beta_r(r, **kwargs_iso)
k = isotropic.K(r, R, **kwargs_iso)
f = isotropic.anisotropy_solution(r, **kwargs_iso)
assert f == 1
print(beta, 'test')
const = Anisotropy(anisotropy_type='const')
kwargs = {'beta': beta}
k_const = const.K(r, R, **kwargs)
print(k, k_const)
npt.assert_almost_equal(k, k_const, decimal=5)
def test_radial_anisotropy(self):
# radial
r = 2.
R = 1.
radial = Anisotropy(anisotropy_type='radial')
kwargs_rad = {}
beta = radial.beta_r(r, **kwargs_rad)
k = radial.K(r, R, **kwargs_rad)
f = radial.anisotropy_solution(r, **kwargs_rad)
assert f == r**2
const = Anisotropy(anisotropy_type='const')
kwargs = {'beta': beta}
print(beta, 'beta')
#kwargs = {'beta': 1}
k_mamon = const.K(r, R, **kwargs)
print(k, k_mamon)
npt.assert_almost_equal(k, k_mamon, decimal=5)
def test_raise(self):
with self.assertRaises(ValueError):
Anisotropy(anisotropy_type='wrong')
with self.assertRaises(ValueError):
ani = Anisotropy(anisotropy_type='Colin')
ani.K(r=1, R=2, r_ani=1)
with self.assertRaises(ValueError):
ani = Anisotropy(anisotropy_type='const')
ani.anisotropy_solution(r=1)
with self.assertRaises(ValueError):
const = Anisotropy(anisotropy_type='const')
kwargs = {'beta': 1}
f_const = const.anisotropy_solution(r=1, **kwargs)