def test_data_save(test_data, tmp_path, WRITE_TOLERANCE):
d_ref = Dataset(test_data[0])
if d_ref.subtype is "":
path_out = tmp_path / d_ref.type
else:
path_out = tmp_path / (d_ref.type + '.' + d_ref.subtype)
d_ref.write(path_out)
d_test = Dataset(path_out)
diff = d_ref.data - d_test.data
max_error = np.max(np.abs(diff))
try:
assert np.array_equal(d_ref.data, d_test.data)
except AssertionError:
pass
if max_error > 0.0:
try:
assert max_error < WRITE_TOLERANCE
print('Arrays are not identical, but max difference: {} is tolerated'.format(max_error))
except AssertionError as e:
raise e
def test_parameters(test_parameters):
dataset = Dataset(test_parameters[0], load=False)
dataset.load_parameters()
for jcampdx in dataset._parameters.values():
with Path(str(jcampdx.path)+'.json').open() as file:
reference = json.load(file)
assert jcampdx.to_dict() == reference
def test_split(test_split_data, tmp_path):
dataset = Dataset(Path(test_split_data[1]) / test_split_data[0]['path'])
if test_split_data[0]['splitter'] == 'SlicePackage':
SlicePackageSplitter().split(dataset, write=True, path_out=tmp_path)
elif test_split_data[0]['splitter'] == 'FG_ECHO':
FrameGroupSplitter('FG_ECHO').split(dataset,
write=True,
path_out=tmp_path)
for ref in test_split_data[0]['results'].values():
ds_split = Dataset(tmp_path / ref['path'])
assert ds_split.shape == tuple(ref['shape'])
def test_data_load(test_data):
dataset = Dataset(test_data[0])
with np.load(str(dataset.path)+'.npz') as data:
try:
assert np.array_equal(dataset.data, data['data'])
except:
print()
def test_ra(test_ra_data):
d1 = Dataset(Path(test_ra_data[1]) / test_ra_data[0]['path'])
core_index = tuple(slice(None) for i in range(d1.encoded_dim))
d2 = Dataset(Path(test_ra_data[1]) / test_ra_data[0]['path'],
random_access=True)
if "slices" in test_ra_data[0].keys():
for s in test_ra_data[0]['slices']:
slice_ = json_to_slice(s)
assert np.array_equal(d1.data[slice_], d2.data[slice_])
else:
# test by single slice - index
for index in np.ndindex(d1.shape[d1.encoded_dim:]):
assert np.array_equal(d1.data[core_index + index],
d2.data[core_index + index])
# test all possible slices
for slice_ in generate_slices(d1.shape[d1.encoded_dim:]):
assert np.array_equal(d1.data[core_index + slice_],
d2.data[core_index + slice_])
def yield_bruker(args):
"""
If the path spectified by args.file is:
1/ Bruker dataset file (2dseq) - function yields its data and properties of the dataset
2/ Directory - function yields data and properties and data of all datasets compliant to the queries
"""
# get location of the spec2nii Bruker properties configuration file
bruker_properties_path = pkg_resources.resource_filename('spec2nii', 'bruker_properties.json')
# get a list of queries to filter datasets
queries = _get_queries(args)
# case of Bruker dataset
if os.path.isfile(args.file):
d = Dataset(args.file, property_files=[bruker_properties_path], parameter_files=['method'])
try:
d.query(queries)
except FilterEvalFalse:
raise ValueError(f'Bruker dataset {d.path} is not suitable for conversion to mrs_nifti')
yield from _proc_dataset(d, args)
# case of folder containing Bruker datasets
elif os.path.isdir(args.file):
# process individual datasets
for dataset in Folder(args.file, dataset_state={
"parameter_files": ['method'],
"property_files": [bruker_properties_path]
}).get_dataset_list_rec():
with dataset as d:
try:
d.query(queries)
except FilterEvalFalse:
continue
yield from _proc_dataset(d, args)
def test_properties(test_properties):
if test_properties:
dataset = Dataset(test_properties[0], load=False, parameter_files=['subject'])
dataset.load_parameters()
dataset.load_properties()
assert dataset.to_dict() == test_properties[1]