def test_mgrid(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
grid_x, grid_y, grid_z = point_list.mgrid(
resolution=sample_logs_mock['resolution'])
assert grid_x[0][0] == pytest.approx([-0.6, -0.6, -0.6, -0.6])
assert grid_x[13][-1] == pytest.approx([0.7, 0.7, 0.7, 0.7])
assert grid_y[0][0] == pytest.approx([0.495, 0.495, 0.495, 0.495])
assert grid_y[13][-1] == pytest.approx([0.601, 0.601, 0.601, 0.601])
assert grid_z[0][0] == pytest.approx([1.498, 1.565, 1.633, 1.700],
abs=0.001)
assert grid_z[13][-1] == pytest.approx([1.498, 1.565, 1.633, 1.700],
abs=0.001)
# Test irreducibility option
# volume scan: all three directions have more than unique point
xyz = [
[0.0, 1.000, 1.001, 2.0, 3.0, 4.0, 0.0, 1.0, 2.0, 3.0, 4.0], # x
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0], # y
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0]
] # z
point_list = PointList(xyz)
grid = point_list.mgrid(resolution=sample_logs_mock['resolution'],
irreducible=True)
assert grid.shape == (3, 5, 2, 2)
# Surface scan on the {vx, vy} plane
xyz[2] = [0] * 11
point_list = PointList(xyz)
grid = point_list.mgrid(resolution=sample_logs_mock['resolution'],
irreducible=True)
assert grid.shape == (2, 5, 2)
# Linear scan on the vx axis
xyz[1] = [0] * 11
point_list = PointList(xyz)
grid = point_list.mgrid(resolution=sample_logs_mock['resolution'],
irreducible=True)
assert grid.shape == (1, 5)
def test_has_overlapping_points(self):
xyz = [[0.0, 1.0, 2.0], [0.0, 0.0, 0.0], [0.0, 0.0, 0.0]]
assert PointList(xyz).has_overlapping_points(
resolution=DEFAULT_POINT_RESOLUTION) is False
xyz = [[0.0, 1.0, 2.0, 2.009], [0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0]]
assert PointList(xyz).has_overlapping_points(
resolution=DEFAULT_POINT_RESOLUTION) is True
def test_is_contained_in(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
subset = np.array(
sample_logs_mock['xyz'])[:, 0:
5] # pick only the first five coordinates
other_list = PointList(subset)
assert other_list.is_contained_in(point_list) is True
assert point_list.is_contained_in(other_list) is False
def test_fuse_with(self):
xyz1 = [[0.0, 1.0, 2.0, 3.0], [0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0]]
xyz2 = [[1.009, 0.995, 2.0, 3.005, 4.0], [0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0]]
point_list = PointList(xyz1)
common = point_list.fuse_with(PointList(xyz2),
resolution=DEFAULT_POINT_RESOLUTION)
assert common.vx == pytest.approx([0.0, 1.0, 2.0, 3.0, 4.0])
def test_intersection(self):
xyz1 = [[0.0, 1.0, 2.0, 3.0], [0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0]]
xyz2 = [[1.009, 0.995, 2.0, 3.005, 4.0], [0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0]]
common = PointList(xyz1).intersection(
PointList(xyz2), resolution=DEFAULT_POINT_RESOLUTION)
assert common.vx == pytest.approx(
[1.0, 2.0, 3.0, 1.009, 0.995, 2.0, 3.005])
def test_aggregate(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs']).aggregate(
PointList(sample_logs_mock['xyz']))
np.testing.assert_equal(
point_list.vx,
sample_logs_mock['xyz'][0] + sample_logs_mock['xyz'][0])
np.testing.assert_equal(
point_list.vy,
sample_logs_mock['xyz'][1] + sample_logs_mock['xyz'][1])
np.testing.assert_equal(
point_list.vz,
sample_logs_mock['xyz'][2] + sample_logs_mock['xyz'][2])
def test_linear_scan_vector(self):
# vx and vy are only one point, within resolution
xyz = [[0, 0, 0.003, 0.004, 0], [0, 1, 2, 3, 4],
[1, 1.001, 1.002, 1.003, 1.009]]
point_list = PointList(xyz)
assert point_list.linear_scan_vector(
resolution=DEFAULT_POINT_RESOLUTION) == pytest.approx([0, 1, 0])
xyz[0][
0] = 0.011 # this point is beyond resolution with the rest of the other vx points
point_list = PointList(xyz)
assert point_list.linear_scan_vector(
resolution=DEFAULT_POINT_RESOLUTION) is None
def test_is_equal_within_resolution(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
# perturb the positions of point_list within resolution
original = np.array(sample_logs_mock['xyz'])
perturbation = (sample_logs_mock['resolution'] /
3.0) * (np.random.random(original.shape) - 0.5)
other_list = PointList(original + perturbation)
assert point_list.is_equal_within_resolution(
other_list, resolution=sample_logs_mock['resolution']) is True
# perturb the very first coordinate too much
perturbation[0][0] += 2 * sample_logs_mock['resolution']
other_list = PointList(original + perturbation)
assert point_list.is_equal_within_resolution(
other_list, resolution=sample_logs_mock['resolution']) is False
def wrapped_function(peaks_data):
r"""
Constructor of `StrainField` objects
Parameters
----------
peaks_data: dict
Returns
-------
~pyrs.dataobjects.fields.StrainField
Examples
--------
"""
# Required arguments
for required in ('subruns', 'wavelength', 'peak_profile', 'background_type', 'd_spacing',
'error_fraction', 'vx', 'vy', 'vz'):
assert required in peaks_data
for key in ('subruns', 'd_reference', 'd_spacing', 'fit_costs', 'vx', 'vy', 'vz'):
if isinstance(peaks_data[key], (list, tuple)):
peaks_data[key] = np.array(peaks_data[key])
for key in peaks_data['native']:
peaks_data['native'][key] = _coerce_to_ndarray(peaks_data['native'][key])
# Default values for optional arguments of the PeakCollection constructor
runnumber = peaks_data.get('runnumber', -1)
peak_collection = PeakCollection(peaks_data['peak_tag'],
peaks_data['peak_profile'], peaks_data['background_type'],
peaks_data['wavelength'],
runnumber=runnumber)
# Back-calculate the peak centers from supplied lattice spacings
centers = 2 * np.rad2deg(np.arcsin(peaks_data['wavelength'] / (2 * peaks_data['d_spacing'])))
# Enter the native parameters in the peak collection
subruns_count = len(peaks_data['subruns'])
peaks_data['native'].update({'PeakCentre': centers})
dtype = get_parameter_dtype(peaks_data['peak_profile'], peaks_data['background_type'])
parameters_value = np.zeros(subruns_count, dtype=dtype)
parameters_error = np.zeros(subruns_count, dtype=dtype)
for parameter_name in peaks_data['native']:
values = peaks_data['native'][parameter_name]
parameters_value[parameter_name] = values
parameters_error[parameter_name] = peaks_data['error_fraction'] * values
# set_peak_fitting_values ensures all required fitting parameters are being passed
peak_collection.set_peak_fitting_values(peaks_data['subruns'],
parameters_value, parameters_error,
peaks_data['fit_costs'])
# set the reference lattice spacing
peak_collection.set_d_reference(1.0, 0.1)
# Point list
point_list = PointList([peaks_data['vx'], peaks_data['vy'], peaks_data['vz']])
return StrainFieldSingle(point_list=point_list, peak_collection=peak_collection)
def test_get_indices(self):
x = [0.0, 0.5, 0.0, 0.5, 0.0, 0.5, 0.0, 0.5] # x
y = [1.0, 1.0, 1.5, 1.5, 1.0, 1.0, 1.5, 1.5] # y
z = [2.0, 2.0, 2.0, 2.0, 2.5, 2.5, 2.5, 2.5] # z
point_list1 = PointList([x, y, z])
# changes a single x-value in the last 4 points of point_list1
x = [0.0, 0.5, 0.0, 0.5, 1.0, 1.5, 1.0, 1.5] # x
y = [1.0, 1.0, 1.5, 1.5, 1.0, 1.0, 1.5, 1.5] # y
z = [2.0, 2.0, 2.0, 2.0, 2.5, 2.5, 2.5, 2.5] # z
point_list2 = PointList([x, y, z])
# reverse order of point_list1
x = [0.5, 0.0, 0.5, 0.0, 0.5, 0.0, 0.5, 0.0]
y = [1.5, 1.5, 1.0, 1.0, 1.5, 1.5, 1.0, 1.0]
z = [2.5, 2.5, 2.5, 2.5, 2.0, 2.0, 2.0, 2.0]
point_list3 = PointList([x, y, z])
# last three sample points of point_list1
x = [0.5, 0.0, 0.5] # x
y = [1.0, 1.5, 1.5] # y
z = [2.5, 2.5, 2.5] # z
point_list4 = PointList([x, y, z])
# indices with self are running numbers
for point_list in (point_list1, point_list2, point_list3, point_list4):
np.testing.assert_equal(point_list.get_indices(point_list),
np.arange(len(point_list)))
# reversing values gets decreasing numbers
for forward, reverse in zip((point_list1, point_list3),
(point_list3, point_list1)):
np.testing.assert_equal(forward.get_indices(reverse),
[7, 6, 5, 4, 3, 2, 1, 0])
# partial matching tests
np.testing.assert_equal(point_list1.get_indices(point_list2),
[0, 1, 2, 3, -1, -1, -1, -1])
np.testing.assert_equal(point_list3.get_indices(point_list2),
[-1, -1, -1, -1, 3, 2, 1, 0])
np.testing.assert_equal(point_list4.get_indices(point_list1),
[5, 6, 7])
np.testing.assert_equal(point_list4.get_indices(point_list2),
[-1, -1, -1])
np.testing.assert_equal(point_list4.get_indices(point_list3),
[2, 1, 0])
def test_is_a_grid(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
assert point_list.is_a_grid(
resolution=sample_logs_mock['resolution']) is False
regular_point_list = point_list.grid_point_list(
resolution=sample_logs_mock['resolution'])
assert regular_point_list.is_a_grid(
resolution=sample_logs_mock['resolution']) is True
def test_cluster(self):
xyz = [[0.0, 1.0, 2.0, 3.0, 1.009, 0.995, 2.0, 3.005, 4.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]]
clusters = PointList(xyz).cluster(resolution=DEFAULT_POINT_RESOLUTION)
for cluster, comparison in zip(clusters,
[[1, 4, 5], [2, 6], [3, 7], [0], [8]]):
assert cluster == pytest.approx(comparison)
def test_coordinates_along_direction(self):
xyz = [[0, 1], [2, 3], [4, 5]]
point_list = PointList(xyz)
for r, index, label in zip(xyz, range(3), ('vx', 'vy', 'vz')):
assert point_list.coordinates_along_direction(
index) == pytest.approx(r)
assert point_list.coordinates_along_direction(
label) == pytest.approx(r)
def test_aggregate_point_list(sample_logs_mock):
point_list = aggregate_point_lists(*[
PointList(sample_logs_mock['logs']) for _ in range(3)
]) # three lists
list_x, list_y, list_z = sample_logs_mock['xyz'][0], sample_logs_mock[
'xyz'][1], sample_logs_mock['xyz'][2]
assert list(point_list.vx) == pytest.approx(list_x + list_x + list_x)
assert list(point_list.vy) == pytest.approx(list_y + list_y + list_y)
assert list(point_list.vz) == pytest.approx(list_z + list_z + list_z)
def test_getattr(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
assert list(point_list.vx) == pytest.approx(sample_logs_mock['xyz'][0])
assert list(point_list.vy) == pytest.approx(sample_logs_mock['xyz'][1])
try:
_ = point_list.dummy # noqa F841
assert False, 'Should not have been able to access attribute'
except AttributeError:
pass # this is the correct behavior
def test_coordinates_irreducible(self):
xyz = [
[0.0, 1.000, 1.001, 2.0, 3.0, 4.0, 0.0, 1.0, 2.0, 3.0, 4.0], # x
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0], # y
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0]
] # z
point_list = PointList(xyz)
# Volume scan
coordinates_irreducible = point_list.coordinates_irreducible(
resolution=DEFAULT_POINT_RESOLUTION)
assert coordinates_irreducible[0] == pytest.approx(
[0.0, 0.0, 0.0]) # check first point
assert coordinates_irreducible[-1] == pytest.approx(
[4.0, 1.0, 1.0]) # check last point
assert coordinates_irreducible.shape[
-1] == 3 # xyz if a volume scan, thus three dimensions
assert coordinates_irreducible == pytest.approx(
np.array(xyz).transpose())
# Surface scan
xyz[2] = [0] * 11
point_list = PointList(xyz)
coordinates_irreducible = point_list.coordinates_irreducible(
resolution=DEFAULT_POINT_RESOLUTION)
assert coordinates_irreducible[0] == pytest.approx(
[0.0, 0.0]) # check first point
assert coordinates_irreducible[-1] == pytest.approx(
[4.0, 1.0]) # check last point
assert coordinates_irreducible.shape[
-1] == 2 # xyz if a surface scan, thus two dimensions
assert coordinates_irreducible == pytest.approx(
np.array(xyz[0:2]).transpose())
# Linear scan
xyz[1] = [0] * 11
point_list = PointList(xyz)
coordinates_irreducible = point_list.coordinates_irreducible(
resolution=DEFAULT_POINT_RESOLUTION)
assert coordinates_irreducible[0] == pytest.approx(
[0.0]) # check first point
assert coordinates_irreducible[-1] == pytest.approx(
[4.0]) # check last point
assert coordinates_irreducible.shape[
-1] == 1 # xyz if a linear scan, thus one dimensions
assert coordinates_irreducible == pytest.approx(
np.array(xyz[0]).reshape((11, 1)))
def test_init(self, sample_logs_mock):
for input_source_type in ('logs', 'xyz'):
input_source = sample_logs_mock[input_source_type]
point_list = PointList(input_source)
# dereference attributes 'vx', 'vy', 'vz'
np.testing.assert_equal(point_list.vx, sample_logs_mock['xyz'][0])
np.testing.assert_equal(point_list.vy, sample_logs_mock['xyz'][1])
np.testing.assert_equal(point_list.vz, sample_logs_mock['xyz'][2])
# dereference by item index
xyz = np.array(sample_logs_mock['xyz'])
assert point_list[5] == pytest.approx(list(xyz[:, 5]))
def test_linspace(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
extents = sample_logs_mock['extents']
vx, vy, vz = point_list.linspace(
resolution=sample_logs_mock['resolution'])
assert [vx[0], vx[-1],
(vx[-1] - vx[0]) / (len(vx) - 1)] == pytest.approx(extents[0])
assert len(vx) == 14
assert [vy[0], vy[-1],
(vy[-1] - vy[0]) / (len(vy) - 1)] == pytest.approx(extents[1])
assert len(vy) == 4 # many values of point_list.vy are repeated
assert [vz[0], vz[-1],
(vz[-1] - vz[0]) / (len(vz) - 1)] == pytest.approx(extents[2])
assert len(vz) == 4 # many values of point_list.vz are repeated
def test_sorted_indices(self):
# The two point lists contain a different number of sample points
point_list = PointList([[0.0, 1.0, 2.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
point_list_other = PointList([[2.0, 0.005], [0.0, 0.0], [0.0, 0.0]])
with pytest.raises(AssertionError) as exception_info:
point_list.sorted_indices(point_list=point_list_other,
resolution=DEFAULT_POINT_RESOLUTION)
assert 'point lists do not contain same number of sample points' in str(
exception_info.value)
# Corner case: one of the point lists contains overlapping points
point_list = PointList([[0.0, 0.009, 1.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]]) # three points, two overlap
with pytest.raises(ValueError) as exception_info:
point_list.sorted_indices(point_list=point_list,
resolution=DEFAULT_POINT_RESOLUTION)
assert 'point lists contains overlapping points' in str(
exception_info.value)
# The two point lists are not equal within resolution
point_list = PointList([[0.0, 1.0, 2.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
point_list_other = PointList([[0.0, 1.1, 2.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
with pytest.raises(ValueError) as exception_info:
point_list.sorted_indices(point_list=point_list_other,
resolution=DEFAULT_POINT_RESOLUTION)
assert 'point lists are not the same, within resolution' in str(
exception_info.value)
# Compare against itself
point_list = PointList([[0.0, 1.0, 2.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
indices = point_list.sorted_indices(
point_list=point_list, resolution=DEFAULT_POINT_RESOLUTION)
np.testing.assert_equal(indices, np.array([0, 1, 2]))
# General case
point_list = PointList([[0.0, 1.0, 2.0], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
point_list_other = PointList([[2.0, 0.005, 1.009], [0.0, 0.0, 0.0],
[0.0, 0.0, 0.0]])
indices = point_list.sorted_indices(
point_list=point_list_other, resolution=DEFAULT_POINT_RESOLUTION)
np.testing.assert_equal(indices, np.array([1, 2, 0]))
def test_grid_point_list(self):
r"""
The regular grid spanned by the three orthonormal vectors is the unit cube, with eight points
The order of the coordinates must follow this nested loop structure:
for vx in ...:
for vy in ...:
for vz in ...:
"""
# Passing the orthonormal vectors along each direction as three points
point_list = PointList([[1, 0, 0], [0, 1, 0], [0, 0, 1]])
other_list = point_list.grid_point_list(
resolution=DEFAULT_POINT_RESOLUTION)
cube_coordinates = [[0., 0., 0.], [0., 0., 1.], [0., 1., 0.],
[0., 1., 1.], [1., 0., 0.], [1., 0., 1.],
[1., 1., 0.], [1., 1., 1.]]
assert np.allclose(cube_coordinates, other_list.coordinates)
def strain_instantiator(name, values, errors, x, y, z):
return StrainField(name,
peak_collection=PeakCollectionLite(name,
strain=values,
strain_error=errors),
point_list=PointList([x, y, z]))
def test_extents(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
for i, extent in enumerate(
point_list.extents(resolution=DEFAULT_POINT_RESOLUTION)):
assert list(extent) == pytest.approx(
sample_logs_mock['extents'][i])
def test_coordinates(self, sample_logs_mock):
point_list = PointList(sample_logs_mock['logs'])
np.testing.assert_allclose(
point_list.coordinates,
np.array(sample_logs_mock['xyz']).transpose())