def test_adaptive_energy_range_builder():
unknown_1 = Spectrum.read_file(io.StringIO(unknown_1_text))
reference_1 = Spectrum.read_file(io.StringIO(reference_1_text))
reference_2 = Spectrum.read_file(io.StringIO(reference_2_text))
fit_energies, fit_energy_indices = AdaptiveEnergyRangeBuilder(
).build_range(unknown_spectrum=unknown_1,
reference_spectrum_seq=[reference_1, reference_2])
assert fit_energies.shape == (2, )
assert fit_energies.iloc[0] == 11765.0
assert fit_energies.iloc[-1] == 11771.0
assert fit_energy_indices.tolist() == [False, True, True, False]
def build(cls, config):
log = logging.getLogger(name=str(cls))
# read section [references]
# support a PRM file such as
# prm = path/to/one.prm
# or
# a list of one or more file globs such as
# arsenic_2_reference_spectra/*.e
# arsenic_3_reference_spectra/*.e
if config.has_section('references'):
if config.has_option('references', 'prm'):
prm_file_path = os.path.expanduser(config.get('references', 'prm'))
reference_spectrum_list, max_cmp, min_cmp = cls.build_reference_spectrum_list_from_prm_file(prm_file_path)
else:
reference_spectrum_list, max_cmp, min_cmp = cls.build_reference_spectrum_list_from_config_file(config)
elif config.has_section('reference_spectra'):
if config.has_option('reference_spectra', 'prm'):
prm_file_path = os.path.expanduser(config.get('reference_spectra', 'pem'))
reference_spectrum_list, max_cmp, min_cmp = cls.build_reference_spectrum_list_from_prm_file(prm_file_path)
else:
raise Exception('section [reference_spectra] is missing required parameter prm')
else:
raise Exception('configuration file is missing section [references]')
if 0 < min_cmp <= max_cmp:
component_count_range = range(min_cmp, max_cmp+1)
logging.info('component count range: {}'.format(component_count_range))
else:
logging.exception('minimum and maximum component counts are not valid')
energy_range = cls.get_energy_range_from_config(config)
# read data files
unknown_spectrum_file_path_list = []
for j, (unknown_spectrum_glob, _) in enumerate(config.items('data')):
log.info('unknown spectrum glob: {}'.format(unknown_spectrum_glob))
glob_pattern_expanded = os.path.expanduser(unknown_spectrum_glob)
unknown_spectrum_file_path_list.extend(glob.glob(glob_pattern_expanded))
log.info('found {} data files'.format(len(unknown_spectrum_file_path_list)))
unknown_spectrum_list = []
for unknown_spectrum_file_path in unknown_spectrum_file_path_list:
log.info('reading data file {}'.format(unknown_spectrum_file_path))
unknown_spectrum = Spectrum.read_file(unknown_spectrum_file_path)
unknown_spectrum_list.append(unknown_spectrum)
fit_task = cls(
reference_spectrum_list=reference_spectrum_list,
unknown_spectrum_list=unknown_spectrum_list,
energy_range_builder=energy_range,
component_count_range=component_count_range
)
return fit_task
def test_arsenic_1(caplog, request):
"""
Test fits for known arsenic data and reference_spectra.
Expect to find PRM, data, and reference files in a directory called 'test_arsenic_fit'.
See also: http://stackoverflow.com/questions/29627341/pytest-where-to-store-expected-data.
:param request: pytest fixture with information about the path to this test file
:return:
"""
caplog.set_level(logging.INFO)
test_arsenic_fit_fp = request.module.__file__
log.info('test_arsenic_fit_fp: {}'.format(test_arsenic_fit_fp))
test_arsenic_fit_dir_path, _ = os.path.splitext(test_arsenic_fit_fp)
#reference_file_path_pattern = os.path.join(test_arsenic_fit_dir_path, 'reference', 'arsenate_*.e')
reference_file_path_pattern = os.path.join(test_arsenic_fit_dir_path,
'reference', '*.e')
data_file_path = os.path.join(test_arsenic_fit_dir_path, 'data',
'OTT3_55_spot0.e')
reference_spectrum_list = [
ReferenceSpectrum.read_file(file_path)
for file_path in glob(reference_file_path_pattern)
]
log.info(reference_spectrum_list)
unknown_spectrum = Spectrum.read_file(data_file_path)
log.info(unknown_spectrum)
task = AllCombinationFitTask(
ls=LinearRegression,
energy_range_builder=AdaptiveEnergyRangeBuilder(),
reference_spectrum_list=reference_spectrum_list,
unknown_spectrum_list=[
unknown_spectrum,
],
best_fits_plot_limit=1,
component_count_range=range(1, 3 + 1))
with tempfile.TemporaryDirectory() as plots_pdf_dp:
task.fit_all(plots_pdf_dp=plots_pdf_dp)
unknown_spectrum_fit = task.fit_table[unknown_spectrum]
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.fit_spectrum_b.shape
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.unknown_spectrum_b.shape
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.residuals.shape
assert 3 == len(unknown_spectrum_fit.best_fit.reference_spectra_seq)
def test_arsenic_1(caplog, request):
"""
Test fits for known arsenic data and reference_spectra.
Expect to find PRM, data, and reference files in a directory called 'test_arsenic_fit'.
See also: http://stackoverflow.com/questions/29627341/pytest-where-to-store-expected-data.
:param request: pytest fixture with information about the path to this test file
:return:
"""
caplog.setLevel(logging.INFO)
test_arsenic_fit_fp = request.module.__file__
log.info('test_arsenic_fit_fp: {}'.format(test_arsenic_fit_fp))
test_arsenic_fit_dir_path, _ = os.path.splitext(test_arsenic_fit_fp)
#reference_file_path_pattern = os.path.join(test_arsenic_fit_dir_path, 'reference', 'arsenate_*.e')
reference_file_path_pattern = os.path.join(test_arsenic_fit_dir_path, 'reference', '*.e')
data_file_path = os.path.join(test_arsenic_fit_dir_path, 'data', 'OTT3_55_spot0.e')
reference_spectrum_list = [
ReferenceSpectrum.read_file(file_path)
for file_path
in glob(reference_file_path_pattern)
]
log.info(reference_spectrum_list)
unknown_spectrum = Spectrum.read_file(data_file_path)
log.info(unknown_spectrum)
task = AllCombinationFitTask(
energy_range_builder=AdaptiveEnergyRangeBuilder(),
reference_spectrum_list=reference_spectrum_list,
unknown_spectrum_list=[unknown_spectrum, ],
component_count_range=range(1, 3+1)
)
task.fit_all()
unknown_spectrum_fit = task.fit_table[unknown_spectrum]
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.fit_spectrum_b.shape
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.unknown_spectrum_b.shape
assert unknown_spectrum_fit.best_fit.interpolant_incident_energy.shape == unknown_spectrum_fit.best_fit.residuals.shape
assert 3 == len(unknown_spectrum_fit.best_fit.reference_spectra_seq)
def build_unknown_spectrum_list_from_config_file(config):
log = logging.getLogger(name=__name__)
unknown_spectrum_file_path_list = []
for j, (unknown_spectrum_glob, _) in enumerate(config.items('data')):
log.info('unknown spectrum glob: {}'.format(unknown_spectrum_glob))
glob_pattern_expanded = os.path.expanduser(unknown_spectrum_glob)
unknown_spectrum_file_path_list.extend(glob(glob_pattern_expanded))
log.info('found {} data files'.format(len(unknown_spectrum_file_path_list)))
unknown_spectrum_list = []
for unknown_spectrum_file_path in unknown_spectrum_file_path_list:
log.info('reading data file {}'.format(unknown_spectrum_file_path))
unknown_spectrum = Spectrum.read_file(unknown_spectrum_file_path)
unknown_spectrum_list.append(unknown_spectrum)
if len(unknown_spectrum_list) == 0:
raise ConfigurationFileError('no spectrum files were found using globs "{}"'.format(config.items('data')))
else:
return unknown_spectrum_list