def testing(self, raw_signal, rfmdl=None):
'''Testing ECG record with trained model.'''
N_original = len(raw_signal)
# default parameters
if rfmdl is None:
rfmdl = self.mdl
PrdRes = []
if valid_signal_value(raw_signal) == False:
raise Exception(
'Input signal value is not valid!(may contain inf)')
feature_extractor = extfeature.ECGfeatures(
raw_signal, random_relation_path=self.SaveTrainingSampleFolder)
N_signal = len(raw_signal)
# Testing...
predict_result = self.testing_with_extractor(rfmdl, N_signal,
feature_extractor)
# Crop the length (SWT may pad zeros behind)
if N_original < len(raw_signal):
del raw_signal[N_original:]
predict_result = predict_result[0:min(len(predict_result
), len(raw_signal))]
PrdRes = predict_result
return PrdRes
def show_drawing(folderpath = os.path.join(\
os.path.dirname(curfilepath),'..','QTdata','QTdata_repo')):
with open(os.path.join(folderpath, 'sel103.txt'), 'r') as fin:
sig = pickle.load(fin)
# sig with 'sig','time'and 'marks'
ECGfv = extfeature.ECGfeatures(sig['sig'])
fv = ECGfv.frompos(3e3)
def test_signal(self, signal, rfmdl=None, MultiProcess='off'):
# test rawsignal
if rfmdl is None:
rfmdl = self.mdl
# Extracting Feature
if MultiProcess == 'off':
FeatureExtractor = extfeature.ECGfeatures(
signal, random_relation_path=self.random_relation_path)
else:
raise StandardError('MultiProcess on is not defined yet!')
# testing
if MultiProcess == 'on':
raise StandardError('MultiProcess on is not defined yet!')
elif MultiProcess == 'off':
record_predict_result = self.test_with_positionlist(
rfmdl, range(0, len(signal)), FeatureExtractor)
return record_predict_result
def testing(self, reclist, rfmdl=None, saveresultfolder=None):
'''Testing ECG record with trained model.'''
# Default parameters
if rfmdl is None:
rfmdl = self.mdl
# Parallel
if conf['ParallelTesting'] == 'True':
MultiProcess = 'on'
else:
MultiProcess = 'off'
# Test all files in reclist
PrdRes = []
for recname in reclist:
# --------------------
# start test time
# --------------------
# load signal data from QTdb.
sig = self.QTloader.load(recname)
# Signal value can not contain inf.
if valid_signal_value(sig['sig']) == False:
continue
# ------------------------
# test lead I
# ------------------------
# original raw_signal
raw_signal = sig['sig']
feature_extractor = extfeature.ECGfeatures(raw_signal)
N_signal = len(raw_signal)
# Testing...
time_rec0 = time.time()
record_predict_result = self.testing_with_extractor(
rfmdl, N_signal, feature_extractor)
# Logging testing time.
time_rec1 = time.time()
log.info('Testing time for %s lead1 is %d seconds', recname,
time_rec1 - time_rec0)
PrdRes.append((recname, record_predict_result))
# ------------------------
# test lead II
# ------------------------
# original raw_signal
raw_signal = sig['sig2']
feature_extractor = extfeature.ECGfeatures(raw_signal)
N_signal = len(raw_signal)
# Testing...
time_rec0 = time.time()
record_predict_result = self.testing_with_extractor(
rfmdl, N_signal, feature_extractor)
# Logging testing time.
time_rec1 = time.time()
log.info('Testing time for %s lead1 is %d seconds', recname,
time_rec1 - time_rec0)
PrdRes.append(('{}_sig2'.format(recname), record_predict_result))
# Save prediction result.
if saveresultfolder is not None:
saveresult_filename = os.path.join(saveresultfolder,
'result_{}'.format(recname))
with open(saveresult_filename, 'w') as fout:
# No detection
if PrdRes is None or len(PrdRes) == 0:
PrdRes = ()
json.dump(PrdRes, fout, indent=4, sort_keys=True)
print 'Saved prediction result to {}'.format(
saveresult_filename)
return PrdRes
def testing(self, reclist, rfmdl=None, saveresultfolder=None):
'''Testing ECG record with trained model.'''
# default parameters
if rfmdl is None:
rfmdl = self.mdl
# Parallel
if conf['ParallelTesting'] == 'True':
MultiProcess = 'on'
else:
MultiProcess = 'off'
# test all files in reclist
PrdRes = []
for recname in reclist:
# --------------------
# start test time
# --------------------
time_rec0 = time.time()
# QT sig data
sig = self.QTloader.load(recname)
# valid signal value:
if valid_signal_value(sig['sig']) == False:
continue
# sigle process
if MultiProcess == 'off':
log.info(
'Multi-process is off, using record-global feature extractor.'
)
FeatureExtractor = extfeature.ECGfeatures(
sig['sig'], random_relation_path=self.random_relation_path)
# ------------------------
# test lead I
# ------------------------
# original rawsig
rawsig = sig['sig']
N_signal = len(rawsig)
# denoise signal
#
if MultiProcess == 'on':
feature_type = conf['feature_type']
if feature_type == 'wavelet':
denoisesig = rawsig
else:
denoisesig = wtdenoise.denoise(rawsig)
denoisesig = denoisesig.tolist()
N_signal = len(denoisesig)
# init
prRes = []
testSamples = []
#
# get prRange:
# test in the same range as expert labels
#
expres = self.QTloader.getexpertlabeltuple(recname)
# Leave Testing Blank Regions in Head&Tail
WindowLen = conf['winlen_ratio_to_fs'] * conf['fs']
Blank_Len = WindowLen / 2 + 1
prRange = range(Blank_Len, N_signal - 1 - Blank_Len)
if conf['QTtest'] == 'FastTest':
TestRegionFolder = os.path.join(projhomepath, 'QTdata',
'QT_TestRegions')
with open(
os.path.join(TestRegionFolder,
'{}_TestRegions.pkl'.format(recname)),
'rb') as fin:
TestRegions = pickle.load(fin)
prRange = []
for region in TestRegions:
prRange.extend(range(region[0], region[1] + 1))
log.info('Testing sample point number is %d for this record',
len(prRange))
if MultiProcess == 'on':
record_predict_result = self.\
test_with_positionlist_multiprocess(\
rfmdl,\
prRange,\
denoisesig,\
origrawsig = rawsig
)
elif MultiProcess == 'off':
record_predict_result = self.\
test_with_positionlist(\
rfmdl,\
prRange,\
FeatureExtractor\
)
#
# prediction result for each record
#
PrdRes.append((recname, record_predict_result))
# end testing time
time_rec1 = time.time()
print 'Testing time for {} is {:.2f} s'.\
format(recname,time_rec1-time_rec0)
log.info('Testing time for %s lead1 is %d seconds', recname,
time_rec1 - time_rec0)
# ------------------------
# test lead II
# ------------------------
# original rawsig
rawsig = sig['sig2']
N_signal = len(rawsig)
# denoise signal
#
if MultiProcess == 'on':
feature_type = conf['feature_type']
if feature_type == 'wavelet':
denoisesig = rawsig
else:
denoisesig = wtdenoise.denoise(rawsig)
denoisesig = denoisesig.tolist()
N_signal = len(denoisesig)
# init
prRes = []
testSamples = []
#
# get prRange:
# test in the same range as expert labels
#
expres = self.QTloader.getexpertlabeltuple(recname)
# Leave Testing Blank Regions in Head&Tail
WindowLen = conf['winlen_ratio_to_fs'] * conf['fs']
Blank_Len = WindowLen / 2 + 1
prRange = range(Blank_Len, N_signal - 1 - Blank_Len)
if conf['QTtest'] == 'FastTest':
TestRegionFolder = os.path.join(projhomepath, 'QTdata',
'QT_TestRegions')
with open(
os.path.join(TestRegionFolder,
'{}_TestRegions.pkl'.format(recname)),
'rb') as fin:
TestRegions = pickle.load(fin)
prRange = []
for region in TestRegions:
prRange.extend(range(region[0], region[1] + 1))
log.info('Testing sample point number is %d for this record',
len(prRange))
if MultiProcess == 'on':
record_predict_result = self.\
test_with_positionlist_multiprocess(\
rfmdl,\
prRange,\
denoisesig,\
origrawsig = rawsig
)
elif MultiProcess == 'off':
record_predict_result = self.\
test_with_positionlist(\
rfmdl,\
prRange,\
FeatureExtractor\
)
#
# prediction result for each record
#
PrdRes.append(('{}_sig2'.format(recname), record_predict_result))
# end testing time
time_rec1 = time.time()
log.info('Testing time for %s lead2 is %d seconds', recname,
time_rec1 - time_rec0)
# save Prediction Result
if saveresultfolder is not None:
# save results
saveresult_filename = os.path.join(saveresultfolder,
'result_{}'.format(recname))
with open(saveresult_filename, 'w') as fout:
# No detection
if PrdRes is None or len(PrdRes) == 0:
PrdRes = ()
json.dump(PrdRes, fout, indent=4, sort_keys=True)
print 'saved prediction result to {}'.format(
saveresult_filename)
return PrdRes
def collectfeaturesforsig(sig,
SaveTrainingSampleFolder,
blankrangelist=None,
recID=None,
allowed_label_list=None,
random_relation_path=None):
'''Label process & convert to feature.'''
#
# parameters:
# blankrangelist : [[l,r],...]
#
# collect training features from sig
#
# init
Extractor = extfeature.ECGfeatures(
sig['sig'], random_relation_path=random_relation_path)
negposlist = []
posposlist = [] # positive position list
labellist = [] # positive label list
tarpos = []
# Including positive samples & negtive samples.
trainingX, trainingy = [], []
# get Expert labels
QTloader = QTdb.QTloader()
# =======================================================
# modified negposlist inside function
# =======================================================
ExpertLabels = QTloader.getexpertlabeltuple(None,
sigIN=sig,
negposlist=negposlist)
# Filtering according to allowed labels
if allowed_label_list is not None:
allowed_expert_labels = filter(
lambda x: x[1] in allowed_label_list, ExpertLabels)
if len(allowed_expert_labels) == 0:
return ([], [])
posposlist, labellist = zip(*allowed_expert_labels)
# ===============================
# convert feature & append to X,y
# Using Map build in function
# ===============================
FV = map(Extractor.frompos, posposlist)
# append to trainging vector
trainingX.extend(FV)
trainingy.extend(labellist)
# add neg samples
Nneg = int(len(negposlist) * conf['negsampleratio'])
# if Number of Neg>0 then add negtive samples
if len(negposlist) == 0 or Nneg <= 0:
print '[In function collect feature] Warning: negtive sample position list length is 0.'
else:
# collect negtive sample features
#
# leave blank for area without labels
#
negposset = set(negposlist)
if blankrangelist is not None:
blklist = []
for pair in blankrangelist:
blklist.extend(range(pair[0], pair[1] + 1))
blkset = set(blklist)
negposset -= blkset
selnegposlist = random.sample(negposset, Nneg)
time_neg0 = time.time()
negFv = map(Extractor.frompos, selnegposlist)
trainingX.extend(negFv)
trainingy.extend(['white'] * Nneg)
print '\nTime for collect negtive samples:{:.2f}s'.format(
time.time() - time_neg0)
# =========================================
# Save sample list
# =========================================
ResultFolder = os.path.join(SaveTrainingSampleFolder,
'TrainingSamples')
# mkdir if not exists
if os.path.exists(ResultFolder) == False:
os.mkdir(ResultFolder)
# -----
# sample_list
# [(pos,label),...]
# -----
sample_list = zip(selnegposlist, len(selnegposlist) * ['white'])
sample_list.extend(ExpertLabels)
if recID is not None:
# save recID sample list
with open(os.path.join(ResultFolder, recID + '.pkl'), 'w') as fout:
pickle.dump(sample_list, fout)
save_mat_filename = os.path.join(ResultFolder, recID + '.mat')
reslist_to_mat(sample_list, mat_filename=save_mat_filename)
return (trainingX, trainingy)