ChanggungPatientNets, ChanggungHealthyNets)
X1 = np.zeros((len(ChanggungHealthyNets), 1)) # healthy
y1 = -1 * np.ones((len(ChanggungHealthyNets), 1)) # label = -1 for healthy
X2 = np.zeros((len(ChanggungPatientNets), 1)) # patient
y2 = np.ones((len(ChanggungPatientNets), 1)) # label = 1 for patients
# Z would be a dict of np.arrays
# The key is the subject name
# The value is a 2d time case X feature dim array
Z = dict()
y3 = dict()
for key in testNets:
Z[key] = np.zeros((len(testNets[key]), 1))
y3[key] = -1 * np.ones((len(testNets[key]), 1))
for c in sig_connections:
normalCList = result_utils.getAllFCAtIdx(c[0], c[1],
ChanggungHealthyNets)
X1 = np.insert(X1, 0, normalCList, axis=1)
patientCList = result_utils.getAllFCAtIdx(c[0], c[1],
ChanggungPatientNets)
X2 = np.insert(X2, 0, patientCList, axis=1)
for key in testNets:
# for each subject
testList = result_utils.getAllFCAtIdx(c[0], c[1], testNets[key])
Z[key] = np.insert(Z[key], 0, testList, axis=1)
X = np.concatenate([X1[:, :-1], X2[:, :-1]])
y = np.concatenate((y1, y2)).ravel()
for key in Z:
Z[key] = Z[key][:, :-1]
# classifier
classifier = svm.SVC(kernel='linear')
trainingNets = copy.deepcopy(ChanggungPatientNets)
leftOutNet = trainingNets[pidx]
trainingNets.remove(leftOutNet)
sig_connections = stats_utils.filter_sigdiff_connections(
trainingNets, ChanggungHealthyNets)
discovery_rates.append(
float(len(sig_connections)) / (atlasobj.count *
(atlasobj.count - 1) / 2))
X1 = np.zeros((len(ChanggungHealthyNets), 1)) # healthy
y1 = -1 * np.ones((len(ChanggungHealthyNets), 1))
X2 = np.zeros((len(trainingNets), 1)) # patient
y2 = np.ones((len(trainingNets), 1))
z = []
for c in sig_connections:
normalCList = result_utils.getAllFCAtIdx(c[0], c[1],
ChanggungHealthyNets)
X1 = np.insert(X1, 0, normalCList, axis=1)
patientCList = result_utils.getAllFCAtIdx(c[0], c[1], trainingNets)
X2 = np.insert(X2, 0, patientCList, axis=1)
z.append(result_utils.getAllFCAtIdx(c[0], c[1], [leftOutNet])[0])
X = np.concatenate([X1[:, :-1], X2[:, :-1]])
y = np.concatenate((y1, y2)).ravel()
print('Sample data shape: %s, label shape: %s' %
(str(X.shape), str(y.shape)))
# train ans score
classifier.fit(X, y)
p = classifier.score(np.array(z, ndmin=2), np.array([1]))
accuracy.append(p)
if p == 1.0:
truePositive += 1
else:
def func(args):
# input parameters
atlasobj = args[0]
ChanggungPatientNets = args[1]
# return results
ret = [None, None, None, None,
None] # discover rate, accuracy, precision, recall, specificity
ChanggungHealthyNets = io_utils.loadRandomDynamicNets(
ChanggungAllFullPath,
atlasobj,
totalNum=len(ChanggungPatientNets),
scanList=os.path.join(ChanggungRootPath, 'normal_scans.txt'))
sig_connections = stats_utils.filter_sigdiff_connections(
ChanggungPatientNets, ChanggungHealthyNets)
ret[0] = float(len(sig_connections)) / (atlasobj.count *
(atlasobj.count - 1) / 2.0)
X1 = np.zeros((len(ChanggungHealthyNets), 1)) # healthy
y1 = -1 * np.ones((len(ChanggungHealthyNets), 1))
X2 = np.zeros((len(ChanggungPatientNets), 1)) # patient
y2 = np.ones((len(ChanggungPatientNets), 1))
for c in sig_connections:
normalCList = result_utils.getAllFCAtIdx(c[0], c[1],
ChanggungHealthyNets)
X1 = np.insert(X1, 0, normalCList, axis=1)
patientCList = result_utils.getAllFCAtIdx(c[0], c[1],
ChanggungPatientNets)
X2 = np.insert(X2, 0, patientCList, axis=1)
X = np.concatenate([X1[:, :-1], X2[:, :-1]])
y = np.concatenate((y1, y2)).ravel()
# classifier
classifier = svm.SVC(kernel='linear')
# leave one out cross validation
accuracy = []
truePositive = 0
falsePositive = 0
trueNegative = 0
falseNegative = 0
loo = model_selection.LeaveOneOut()
for trainIdx, testIdx in loo.split(X, y):
classifier.fit(X[trainIdx, :], y[trainIdx])
accuracy.append(classifier.score(X[testIdx, :], y[testIdx]))
p = classifier.predict(X[testIdx, :])[0]
if p == 1 and y[testIdx] == 1:
truePositive += 1
elif p == -1 and y[testIdx] == -1:
trueNegative += 1
elif p == 1 and y[testIdx] == -1:
falsePositive += 1
else:
falseNegative += 1
ret[1] = np.mean(accuracy)
precision = float(truePositive) / (truePositive + falsePositive)
recall = float(truePositive) / (truePositive + falseNegative)
specificity = float(trueNegative) / (trueNegative + falsePositive)
ret[2] = precision
ret[3] = recall
ret[4] = specificity
return ret
'SCI_incomplete_subjects.txt'))
ChanggungHealthyNets = io_utils.loadSpecificNets(
mmdps_locale.ChanggungAllFullPath,
atlasobj,
subjectList=os.path.join(mmdps_locale.ChanggungRootPath,
'normal_subjects.txt'))
# prepare all training
X1 = np.zeros((len(ChanggungHealthyNets), 1)) # healthy
y1 = -1 * np.ones((len(ChanggungHealthyNets), 1))
X2 = np.zeros((len(ChanggungPatientNets), 1)) # patient
y2 = np.ones((len(ChanggungPatientNets), 1))
for xidx in range(atlasobj.count):
print('xidx = %d' % xidx)
for yidx in range(xidx + 1, atlasobj.count):
normalCList = result_utils.getAllFCAtIdx(xidx, yidx,
ChanggungHealthyNets)
X1 = np.insert(X1, 0, normalCList, axis=1)
patientCList = result_utils.getAllFCAtIdx(xidx, yidx,
ChanggungPatientNets)
X2 = np.insert(X2, 0, patientCList, axis=1)
X = np.concatenate([X1[:, :-1], X2[:, :-1]])
y = np.concatenate((y1, y2)).ravel()
print('Sample data shape: %s, label shape: %s' %
(str(X.shape), str(y.shape)))
# classifier
classifier = svm.SVC(kernel='linear')
# leave one out cross validation
accuracy = []
truePositive = 0