def get_corr_files(ensemble_name, search_dir):
file_list_infos = dict()
for root, dirs, files in os.walk(search_dir, topdown=True):
for filename in files:
full_filename = os.path.join(root, filename)
try:
file_type = sigmond.getFileID(full_filename)
except ValueError:
continue
if file_type != sigmond.FileType.Correlator:
continue
base, ext = os.path.splitext(full_filename)
suffix = int(ext[1:])
if base in file_list_infos:
if suffix < file_list_infos[base][0]:
file_list_infos[base][0] = suffix
elif suffix > file_list_infos[base][1]:
file_list_infos[base][1] = suffix
else:
file_list_infos[base] = [suffix, suffix]
file_list_infos = [sigmond.FileListInfo(stub, suffix[0], suffix[1], False)
for stub, suffix in file_list_infos.items()]
mcobs_xml = ET.Element("MCObservables")
corr_data_xml = ET.SubElement(mcobs_xml, "BLCorrelatorData")
for file_list_info in file_list_infos:
corr_data_xml.append(file_list_info.xml())
mcobs_xml_handler = sigmond.XMLHandler()
mcobs_xml_handler.set_from_string(ET.tostring(mcobs_xml))
ensemble_info = sigmond.MCEnsembleInfo(ensemble_name, 'ensembles.xml')
sampling_info = sigmond.MCSamplingInfo()
bins_info = sigmond.MCBinsInfo(ensemble_info)
bins_info.addOmissions(defs.omissions[ensemble_name])
obs_get_handler = sigmond.MCObsGetHandler(mcobs_xml_handler, bins_info, sampling_info)
obs_handler = sigmond.MCObsHandler(obs_get_handler, False)
corr_handler = sigmond.BLCorrelatorDataHandler(file_list_infos, set(), set(), ensemble_info)
corr_files = dict()
for corr in corr_handler.getCorrelatorSet():
channel = get_channel(corr)
if channel not in corr_files:
corr_files[channel] = dict()
corr_files[channel]['data_files'] = list()
corr_files[channel]['correlators'] = list()
corr_files[channel]['data_files'].append(corr_handler.getFileName(corr))
corr_files[channel]['correlators'].append(corr)
return corr_files
def get_corr_files(ensemble_name, search_dir):
print("finding files...", end='')
file_list_infos = dict()
for root, dirs, files in os.walk(search_dir, topdown=True):
for filename in files:
full_filename = os.path.join(root, filename)
try:
file_type = sigmond.getFileID(full_filename)
except ValueError:
continue
if file_type != sigmond.FileType.Correlator:
continue
base, ext = os.path.splitext(full_filename)
suffix = int(ext[1:])
if base in file_list_infos:
if suffix < file_list_infos[base][0]:
file_list_infos[base][0] = suffix
elif suffix > file_list_infos[base][1]:
file_list_infos[base][1] = suffix
else:
file_list_infos[base] = [suffix, suffix]
print("done")
file_list_infos = [
sigmond.FileListInfo(stub, suffix[0], suffix[1], False)
for stub, suffix in file_list_infos.items()
]
mcobs_xml = ET.Element("MCObservables")
corr_data_xml = ET.SubElement(mcobs_xml, "BLCorrelatorData")
for file_list_info in file_list_infos:
corr_data_xml.append(file_list_info.xml())
mcobs_xml_handler = sigmond.XMLHandler()
mcobs_xml_handler.set_from_string(ET.tostring(mcobs_xml))
ensemble_info = sigmond.MCEnsembleInfo(ensemble_name, 'ensembles.xml')
sampling_info = sigmond.MCSamplingInfo()
bins_info = sigmond.MCBinsInfo(ensemble_info)
bins_info.addOmissions(defs.omissions[ensemble_name])
obs_get_handler = sigmond.MCObsGetHandler(mcobs_xml_handler, bins_info,
sampling_info)
obs_handler = sigmond.MCObsHandler(obs_get_handler, False)
print("searching through correlators...", end='')
corr_handler = sigmond.BLCorrelatorDataHandler(file_list_infos, set(),
set(), ensemble_info)
corr_files = dict()
for corr in corr_handler.getCorrelatorSet():
corr_file_name = corr_handler.getFileName(corr)
if corr.isBackwards():
corr.setForwards()
corr_opposite = sigmond.CorrelatorInfo(corr.getSource(),
corr.getSink())
if corr_opposite in corr_files:
corr_files[corr_opposite][1] = corr_file_name
else:
corr_files[corr] = (corr_file_name, None)
print("done")
return corr_files
def find_data(ensemble_name, search_dir):
ensemble_info = sigmond.MCEnsembleInfo(ensemble_name, 'ensembles.xml')
# search for data files
bin_files = set()
smp_files = set()
print(f"Searching for correlators and time separations in {search_dir}")
for root, dirs, files in os.walk(search_dir, topdown=True):
for filename in files:
full_filename = os.path.join(root, filename)
try:
file_type = sigmond.getFileID(full_filename)
except ValueError:
continue
if file_type == sigmond.FileType.Bins:
bin_files.add(full_filename)
elif file_type == sigmond.FileType.Samplings:
smp_files.add(full_filename)
else:
continue
if not BINS and bin_files:
print("ignoring bin files found")
elif BINS and smp_files:
print("ignoring smp files found")
# create sigmond handlers
mcobs_xml = ET.Element("MCObservables")
if BINS:
bins_data_xml = ET.SubElement(mcobs_xml, "BinData")
for bin_file in bin_files:
ET.SubElement(bins_data_xml, "FileName").text = bin_file
else:
smp_data_xml = ET.SubElement(mcobs_xml, "SamplingData")
for smp_file in smp_files:
ET.SubElement(smp_data_xml, "FileName").text = smp_file
mcobs_xml_handler = sigmond.XMLHandler()
mcobs_xml_handler.set_from_string(ET.tostring(mcobs_xml))
sampling_info = sigmond.MCSamplingInfo()
bins_info = sigmond.MCBinsInfo(ensemble_info)
bins_info.addOmissions(defs.omissions[ensemble_name])
obs_get_handler = sigmond.MCObsGetHandler(mcobs_xml_handler, bins_info,
sampling_info)
obs_handler = sigmond.MCObsHandler(obs_get_handler, False)
if BINS:
data_handler = sigmond.BinsGetHandler(bins_info, bin_files)
data_size = bins_info.getNumberOfBins()
else:
data_handler = sigmond.SamplingsGetHandler(bins_info, sampling_info,
smp_files)
data_size = sampling_info.getNumberOfReSamplings(bins_info) + 1
obs_infos = data_handler.getKeys()
data_dict = dict()
max_level = dict()
max_time = dict()
data_size
for obs_info in obs_infos:
if obs_info.isImaginaryPart():
continue
operator = obs_info.getCorrelatorSourceInfo().getGenIrrep()
channel = get_channel(operator)
if channel not in data_dict:
data_dict[channel] = dict()
if channel not in max_level:
max_level[channel] = 0
if channel not in max_time:
max_time[channel] = 0
level = operator.getIDIndex()
if level not in data_dict[channel]:
data_dict[channel][level] = dict()
max_level[channel] = max(max_level[channel], level)
time = obs_info.getCorrelatorTimeIndex()
#print(f"{channel!r} - {level} - {time}")
if time in data_dict[channel][level]:
print("another problem")
exit()
max_time[channel] = max(max_time[channel], time)
if BINS:
data = obs_handler.getBins(obs_info).array()
else:
data = obs_handler.getFullAndSamplingValues(
obs_info, sampling_info.getSamplingMode()).array()
data_dict[channel][level][time] = data
final_data_dict = dict()
for channel, levels in data_dict.items():
data_arr = np.zeros(
(data_size, max_level[channel] + 1, max_time[channel] + 1),
dtype=np.float64)
for level, times in levels.items():
for time, data in times.items():
data_arr[:, level, time] = data
final_data_dict[channel] = data_arr
return final_data_dict