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_2(caplog, request):
"""
Test fits for a single reference against all 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.
This test is not reliable. Fix it.
: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, 'reference', 'arsenate_aqueous_avg_als_cal.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 = reference_spectrum_list[0]
log.info(unknown_spectrum)
task = AllCombinationFitTask(
ls=LinearRegression,
energy_range_builder=FixedEnergyRangeBuilder(energy_start=11850.0,
energy_stop=12090.0),
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]
best_fit_ref_count = len(
unknown_spectrum_fit.best_fit.reference_spectra_seq)
assert 2 <= best_fit_ref_count <= 3
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_reference_spectrum_list_from_config_file(config):
"""
Read reference spectrum file glob(s) from configuration file to create
and return a list of ReferenceSpectrum instances.
:param config: configparser instance
:return: list of ReferenceSpectrum instances
"""
log = logging.getLogger(name=__name__)
references = config.items('references')
log.debug(references)
reference_spectrum_list, _ = ReferenceSpectrum.read_all(
[os.path.expanduser(reference_file_glob) for reference_file_glob, _ in references])
if len(reference_spectrum_list) == 0:
raise ConfigurationFileError('no reference spectrum files were found using globs "{}"'.format(references))
else:
return reference_spectrum_list
def build_reference_spectrum_list_from_config_file(cls, config):
"""
Read reference spectrum file glob(s) from configuration file to create
a list of ReferenceSpectrum instances, maximum component count, and
minimum component count.
:param config: configparser instance
:return: list of ReferenceSpectrum instances
"""
reference_spectrum_list, reference_spectrum_file_glob_list = ReferenceSpectrum.read_all(
[reference_file_glob for reference_file_glob, _ in config.items('references')]
)
if len(reference_spectrum_file_glob_list) == 0:
logging.exception(
'no reference spectrum file paths or patterns were found in section [references] of configuration file {}'.format(
config
)
)
elif len(reference_spectrum_list) == 0:
logging.exception('no reference spectrum files were found')
else:
# everything is ok
pass
if not config.has_option('fit', 'maximum_component_count'):
logging.exception(
'required parameter maximum_component_count is missing from section [fit] in configuration file {}'.format(
config
)
)
elif not config.has_option('fit', 'minimum_component_count'):
logging.exception(
'required parameter minimum_component_count is missing from section [fit] in configuration file {}'.format(
config
)
)
else:
maximum_component_count = config.getint('fit', 'maximum_component_count')
minimum_component_count = config.getint('fit', 'minimum_component_count')
return reference_spectrum_list, maximum_component_count, minimum_component_count
def build_reference_spectrum_list_from_prm_file(cls, prm_file_path):
"""
Read a PRM file to create a list of ReferenceSpectrum
instances, maximum component count, and minimum component
count from a PRM file.
:param prm_file_path:
:return:
list of ReferenceSpectrum instances
maximum component count
minimum component count
"""
reference_spectrum_list = []
log.info('reading PRM file {}'.format(prm_file_path))
prm = PRM.read_prm(prm_file_path)
# read reference files
for i, fp in enumerate(prm.reference_file_path_list):
log.info('reading reference file {}: {}'.format(i, fp))
reference_spectrum = ReferenceSpectrum.read_file(fp)
reference_spectrum_list.append(reference_spectrum)
return reference_spectrum_list, prm.nb_component_max, prm.nb_component_min
def build_reference_spectrum_list_from_prm_file(prm_file_path):
"""
Read a PRM file to create a list of ReferenceSpectrum
instances, maximum component count, and minimum component
count from a PRM file.
:param prm_file_path:
:return:
list of ReferenceSpectrum instances
maximum component count
minimum component count
"""
log = logging.getLogger(name=__file__)
reference_spectrum_list = []
log.info('reading PRM file {}'.format(prm_file_path))
prm = PRM.read_prm(prm_file_path)
# read reference files
for i, fp in enumerate(prm.reference_file_path_list):
log.info('reading reference file {}: {}'.format(i, fp))
reference_spectrum = ReferenceSpectrum.read_file(fp)
reference_spectrum_list.append(reference_spectrum)
return reference_spectrum_list, prm.nb_component_max, prm.nb_component_min
def build_reference_spectrum_list_from_config_prm_section(config):
log = logging.getLogger(name=__name__)
max_component_count = int(_get_required_config_value(config, 'prm', 'NBCompoMax'))
min_component_count = int(_get_required_config_value(config, 'prm', 'NBCompoMin'))
reference_spectrum_list = [
ReferenceSpectrum.read_file(file_path_or_buffer=option_name)
for option_name, option_value
in _get_required_config_value_list(config, 'prm')
if len(option_value) == 0]
log.debug('NBCompoMax: %d', max_component_count)
log.debug('NBCompoMin: %d', min_component_count)
log.debug('Reference list length:\n %d', len(reference_spectrum_list))
if min_component_count <= 0:
raise ConfigurationFileError('NBCompoMin must be greater than zero, not "{}"'.format(min_component_count))
elif max_component_count <= 0:
raise ConfigurationFileError('NBCompoMax must be greater than zero, not "{}"'.format(max_component_count))
elif min_component_count > max_component_count:
raise ConfigurationFileError(
'NBCompoMin "{}" is greater than NBCompoMax "{}"'.format(min_component_count, max_component_count))
else:
return max_component_count, min_component_count, reference_spectrum_list
