def test__grid(
self,
image_7x7,
noise_map_7x7,
sub_mask_2d_7x7,
grid_2d_7x7,
sub_grid_2d_7x7,
grid_2d_iterate_7x7,
):
masked_imaging_7x7 = abstract_dataset.AbstractDataset(
data=aa.Array1D.manual_native(array=[1.0], pixel_scales=1.0),
noise_map=aa.Array1D.manual_native(array=[1.0], pixel_scales=1.0),
settings=abstract_dataset.AbstractSettingsDataset(),
)
assert isinstance(masked_imaging_7x7.grid, aa.Grid1D)
masked_image_7x7 = aa.Array2D.manual_mask(
array=image_7x7.native, mask=sub_mask_2d_7x7.mask_sub_1)
masked_noise_map_7x7 = aa.Array2D.manual_mask(
array=noise_map_7x7.native, mask=sub_mask_2d_7x7.mask_sub_1)
masked_imaging_7x7 = abstract_dataset.AbstractDataset(
data=masked_image_7x7,
noise_map=masked_noise_map_7x7,
settings=abstract_dataset.AbstractSettingsDataset(),
)
assert isinstance(masked_imaging_7x7.grid, aa.Grid2D)
assert (masked_imaging_7x7.grid.binned == grid_2d_7x7).all()
assert (masked_imaging_7x7.grid.slim == sub_grid_2d_7x7).all()
masked_imaging_7x7 = abstract_dataset.AbstractDataset(
data=masked_image_7x7,
noise_map=masked_noise_map_7x7,
settings=abstract_dataset.AbstractSettingsDataset(
grid_class=aa.Grid2DIterate),
)
assert isinstance(masked_imaging_7x7.grid, aa.Grid2DIterate)
assert (masked_imaging_7x7.grid.binned == grid_2d_iterate_7x7).all()
masked_imaging_7x7 = abstract_dataset.AbstractDataset(
data=masked_image_7x7,
noise_map=masked_noise_map_7x7,
settings=abstract_dataset.AbstractSettingsDataset(
grid_class=aa.Grid2DInterpolate, pixel_scales_interp=1.0),
)
grid = aa.Grid2DInterpolate.from_mask(mask=sub_mask_2d_7x7,
pixel_scales_interp=1.0)
assert isinstance(masked_imaging_7x7.grid, aa.Grid2DInterpolate)
assert (masked_imaging_7x7.grid == grid).all()
assert (masked_imaging_7x7.grid.grid_interp == grid.grid_interp).all()
assert (masked_imaging_7x7.grid.vtx == grid.vtx).all()
assert (masked_imaging_7x7.grid.wts == grid.wts).all()
def test__image_and_noise_are_values__signal_to_noise_is_absolute_image_value_over_noise(
self):
array = aa.array.manual_2d([[-1.0, 2.0], [3.0, -4.0]])
noise_map = aa.array.manual_2d([[10.0, 10.0], [30.0, 4.0]])
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.potential_chi_squared_map.in_2d == np.array(
[[0.1**2.0, 0.2**2.0], [0.1**2.0, 1.0**2.0]])).all()
assert dataset.potential_chi_squared_max == 1.0
def test__image_and_noise_are_values__signal_to_noise_is_ratio_of_each(
self):
array = aa.array.manual_2d([[1.0, 2.0], [3.0, 4.0]])
noise_map = aa.array.manual_2d([[10.0, 10.0], [30.0, 4.0]])
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.signal_to_noise_map.in_2d == np.array([[0.1, 0.2],
[0.1,
1.0]])).all()
assert dataset.signal_to_noise_max == 1.0
def test__same_as_above__but_image_has_negative_values__replaced_with_zeros(
self):
array = aa.array.manual_2d([[-1.0, 2.0], [3.0, -4.0]])
noise_map = aa.array.manual_2d([[10.0, 10.0], [30.0, 4.0]])
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.signal_to_noise_map.in_2d == np.array([[0.0, 0.2],
[0.1,
0.0]])).all()
assert dataset.signal_to_noise_max == 0.2
def test__potential_chi_squared_map__image_and_noise_are_values__signal_to_noise_is_absolute_image_value_over_noise(
self, ):
array = aa.Array2D.manual_native([[-1.0, 2.0], [3.0, -4.0]],
pixel_scales=1.0)
noise_map = aa.Array2D.manual_native([[10.0, 10.0], [30.0, 4.0]],
pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.potential_chi_squared_map.native == np.array(
[[0.1**2.0, 0.2**2.0], [0.1**2.0, 1.0**2.0]])).all()
assert dataset.potential_chi_squared_max == 1.0
def test__inverse_noise_is_one_over_noise(self):
array = aa.Array2D.manual_native([[1.0, 2.0], [3.0, 4.0]],
pixel_scales=1.0)
noise_map = aa.Array2D.manual_native([[1.0, 2.0], [4.0, 8.0]],
pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.inverse_noise_map.native == np.array([[1.0, 0.5],
[0.25,
0.125]])).all()
def test__grids_change_sub_size_using_settings(self, image_7x7,
noise_map_7x7):
masked_dataset = abstract_dataset.AbstractDataset(
data=image_7x7,
noise_map=noise_map_7x7,
settings=abstract_dataset.AbstractSettingsDataset(
sub_size=1, sub_size_inversion=1),
)
assert masked_dataset.grid.mask.sub_size == 1
assert masked_dataset.grid_inversion.mask.sub_size == 1
masked_dataset = abstract_dataset.AbstractDataset(
data=image_7x7,
noise_map=noise_map_7x7,
settings=abstract_dataset.AbstractSettingsDataset(
sub_size=2, sub_size_inversion=2),
)
assert masked_dataset.grid.mask.sub_size == 2
assert masked_dataset.grid_inversion.mask.sub_size == 2
def test__signal_to_noise_map__image_and_noise_are_values__signal_to_noise_is_ratio_of_each(
self, ):
array = aa.Array2D.manual_native([[1.0, 2.0], [3.0, 4.0]],
pixel_scales=1.0)
noise_map = aa.Array2D.manual_native([[10.0, 10.0], [30.0, 4.0]],
pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.signal_to_noise_map.native == np.array([[0.1, 0.2],
[0.1,
1.0]])).all()
assert dataset.signal_to_noise_max == 1.0
def test__dataset_takes_structures_of_different_formats(self):
array = aa.Array1D.manual_native([1.0, 2.0], pixel_scales=1.0)
noise_map = aa.Array1D.manual_native([1.0, 3.0], pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.data.native == np.array([1.0, 2.0])).all()
assert (dataset.noise_map.native == np.array([1.0, 3.0])).all()
array = aa.Array2D.manual_native([[1.0, 2.0], [3.0, 4.0]],
pixel_scales=1.0)
noise_map = aa.Array2D.manual_native([[1.0, 2.0], [3.0, 5.0]],
pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.data.native == np.array([[1.0, 2.0], [3.0,
4.0]])).all()
assert (dataset.noise_map.native == np.array([[1.0, 2.0],
[3.0, 5.0]])).all()
def test__signal_to_noise_map__same_as_above__but_image_has_negative_values__replaced_with_zeros(
self, ):
array = aa.Array2D.manual_native([[-1.0, 2.0], [3.0, -4.0]],
pixel_scales=1.0)
noise_map = aa.Array2D.manual_native([[10.0, 10.0], [30.0, 4.0]],
pixel_scales=1.0)
dataset = abstract_dataset.AbstractDataset(data=array,
noise_map=noise_map)
assert (dataset.signal_to_noise_map.native == np.array([[0.0, 0.2],
[0.1,
0.0]])).all()
assert dataset.signal_to_noise_max == 0.2
def test__new_imaging_with_arrays_trimmed_via_kernel_shape(self):
data = aa.Array2D.full(fill_value=20.0,
shape_native=(3, 3),
pixel_scales=1.0)
data[4] = 5.0
noise_map_array = aa.Array2D.full(fill_value=5.0,
shape_native=(3, 3),
pixel_scales=1.0)
noise_map_array[4] = 2.0
dataset = abstract_dataset.AbstractDataset(data=data,
noise_map=noise_map_array)
dataset_trimmed = dataset.trimmed_after_convolution_from(
kernel_shape=(3, 3))
assert (dataset_trimmed.data.native == np.array([[5.0]])).all()
assert (dataset_trimmed.noise_map.native == np.array([[2.0]])).all()