def set_fisher_info(self):
for e in xrange(self.data.shape[0]):
if np.all(self.data[e][:1000] == 0):
continue
name = 'fisher_e' + str(e)
value = pyeeg.fisher_info(W = self.svd_embed_seq['svd_embed_seq_e' + str(e)])
self.features[name] = value
def features(mat):
Kmax = 5
Tau = 4
DE = 10
M = 10
R = 0.3
Band = np.arange(1, 86)
Fs = 173
DFA = pyeeg.dfa(mat)
HFD = pyeeg.hfd(mat, Kmax)
SVD_Entropy = pyeeg.svd_entropy(mat, Tau, DE)
Fisher_Information = pyeeg.fisher_info(mat, Tau, DE)
PFD = pyeeg.pfd(mat)
sleep(0.01)
return (DFA, HFD, SVD_Entropy, Fisher_Information, PFD)
def features(mat):
Kmax = 5
Tau = 4
DE = 10
M = 10
R = 0.3
Band = np.arange(1, 86)
Fs = 173
DFA = pyeeg.dfa(mat)
HFD = pyeeg.hfd(mat, Kmax)
SVD_Entropy = pyeeg.svd_entropy(mat, Tau, DE)
Fisher_Information = pyeeg.fisher_info(mat, Tau, DE)
#ApEn = pyeeg.ap_entropy(mat, M, R) # very slow
PFD = pyeeg.pfd(mat)
Spectral_Entropy = pyeeg.spectral_entropy(mat, Band, Fs, Power_Ratio=None)
sleep(0.01)
return (DFA, HFD, SVD_Entropy, Fisher_Information, PFD, Spectral_Entropy)
def advanced_statistics(signal, fs=128):
K_boundary = 10 # to be tuned
t_fisher = 12 # to be tuned
d_fisher = 40 # to be tuned
features_num = 11
threshold = 0.0009
advanced_stats = np.zeros((signal.shape[0], features_num))
print("Gathering advanced statistics...")
for i in tqdm((np.arange(signal.shape[0]))):
feat_array = np.array([
pyeeg.fisher_info(signal[i, :], t_fisher, d_fisher),
pyeeg.pfd(signal[i, :]),
pyeeg.dfa(signal[i, :]),
pyeeg.hfd(signal[i, :], K_boundary),
np.sum((abs(signal[i, :])**(-0.3)) > 20),
np.sum((abs(signal[i, :])) > threshold),
np.std(abs(signal[i, :])**(0.05)),
np.sqrt(np.mean(np.power(np.diff(signal[i, :]), 2))),
np.mean(np.abs(np.diff(signal[i, :]))),
np.mean(signal[i, :]**5),
np.sum(signal[i, :]**2)
])
advanced_stats[i, :] = feat_array
return advanced_stats
SPECT_ENT_BANDS = 2 ** np.arange(0,8)/2
fun_to_test = [
{"times":100,"name":"hfd", "is_original":True,"fun": lambda x: pyeeg.hfd(x,2**3)},
{"times":100,"name":"hfd", "is_original":False,"fun": lambda x: univ.hfd(x,2**3)},
{"times":100,"name":"hjorth", "is_original":True,"fun": lambda x: pyeeg.hjorth(x)},
{"times":100,"name":"hjorth", "is_original":False,"fun": lambda x: univ.hjorth(x)},
{"times":100,"name":"pfd", "is_original":True, "fun":lambda x: pyeeg.pfd(x)},
{"times":100,"name":"pfd", "is_original":False, "fun":lambda x: pyeeg.pfd(x)},
{"times":2,"name":"samp_ent", "is_original":True, "fun":lambda x: pyeeg.samp_entropy(x,2,1.5)},
{"times":10,"name":"samp_ent", "is_original":False, "fun":lambda x: univ.samp_entropy(x,2,1.5,relative_r=False)},
{"times":2,"name":"ap_ent", "is_original":True, "fun":lambda x: pyeeg.ap_entropy(x,2,1.5)},
{"times":10,"name":"ap_ent", "is_original":False, "fun":lambda x: univ.ap_entropy(x,2,1.5)},
{"times":10,"name":"svd_ent", "is_original":True, "fun":lambda x: pyeeg.svd_entropy(x,2,3)},
{"times":100,"name":"svd_ent", "is_original":False, "fun":lambda x: univ.svd_entropy(x,2,3)},
{"times":10,"name":"fisher_info", "is_original":True, "fun":lambda x: pyeeg.fisher_info(x,2,3)},
{"times":100, "name":"fisher_info", "is_original":False, "fun":lambda x: univ.fisher_info(x,2,3)},
{"times":100,"name":"spectral_entropy", "is_original":True, "fun":lambda x: pyeeg.spectral_entropy(x,SPECT_ENT_BANDS,256)},
{"times":100, "name":"spectral_entropy", "is_original":False, "fun":lambda x: univ.spectral_entropy(x,256, SPECT_ENT_BANDS)},
]
def make_one_rep():
ldfs = []
for n in range(MIN_EPOCH_N, MAX_EPOCH_N + 1, EPOCH_STEP):
a = numpy.random.normal(size=n)
for fun in fun_to_test:
f = lambda : fun["fun"](a)
t=Timer(f)
numb = fun["times"]
def feature_wave(self, toolName=None, Fs=256):
if (toolName == None):
print('please select a tool')
return
if toolName in self.FeatureSet.dict[0]:
index = self.FeatureSet.dict[0][toolName]
else:
index = -1
print(toolName)
if toolName == 'DWT':
answer_train = DWT(self.DataSet.trainSet_data[0], 'db4')
answer_test = DWT(self.DataSet.testSet_data[0], 'db4')
print('DWT feature extraction succeed db4')
elif toolName == 'hurst':
answer_train = [
pyeeg.hurst(i) for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.hurst(i) for i in self.DataSet.testSet_data[0]
]
print('hurst feature extraction succeed')
elif toolName == 'dfa':
answer_train = [
pyeeg.dfa(i, L=[4, 8, 16, 32, 64])
for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.dfa(i, L=[4, 8, 16, 32, 64])
for i in self.DataSet.testSet_data[0]
]
print('dfa feature extraction succeed')
elif toolName == 'fisher_info':
answer_train = [
pyeeg.fisher_info(i, 2, 20)
for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.fisher_info(i, 2, 20)
for i in self.DataSet.testSet_data[0]
]
print('fisher_info feature extraction succeed')
elif toolName == 'svd_entropy':
answer_train = [
pyeeg.svd_entropy(i, 2, 20)
for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.svd_entropy(i, 2, 20)
for i in self.DataSet.testSet_data[0]
]
print('svd_entropy feature extraction succeed')
elif toolName == 'spectral_entropy':
bandlist = [0.5, 4, 7, 12, 30, 100]
answer_train = [
pyeeg.spectral_entropy(i, bandlist, Fs)
for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.spectral_entropy(i, bandlist, Fs)
for i in self.DataSet.testSet_data[0]
]
print('spectral_entropy feature extraction succeed')
elif toolName == 'hjorth':
# 得到两个量 第一个是 mobility 第二个是 complexity
answer_train = [
pyeeg.hjorth(i) for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.hjorth(i) for i in self.DataSet.testSet_data[0]
]
answer_train = np.array(answer_train)
answer_test = np.array(answer_test)
for i in answer_train:
i[1] = i[1] / 100
for i in answer_test:
i[1] = i[1] / 100
#只取Mobility
answer_train = np.array(answer_train[:, 0])
answer_test = np.array(answer_test[:, 0])
print('hjorth feature extraction succeed')
elif toolName == 'hfd':
answer_train = [
pyeeg.hfd(i, 8) for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.hfd(i, 8) for i in self.DataSet.testSet_data[0]
]
print('hfd feature extraction succeed')
elif toolName == 'pfd':
answer_train = [
pyeeg.pfd(i) for i in self.DataSet.trainSet_data[0]
]
answer_test = [pyeeg.pfd(i) for i in self.DataSet.testSet_data[0]]
print('pfd feature extraction succeed')
elif toolName == 'bin_power':
bandlist = [0.5, 4, 7, 12] #,30,100]
answer_train = [
pyeeg.bin_power(i, bandlist, Fs)
for i in self.DataSet.trainSet_data[0]
]
answer_test = [
pyeeg.bin_power(i, bandlist, Fs)
for i in self.DataSet.testSet_data[0]
]
print('bin_power feature extraction succeed')
else:
print('does not have this kind of mode')
answer_train = np.array(answer_train)
answer_train = answer_train.reshape(len(answer_train), -1)
answer_test = np.array(answer_test)
answer_test = answer_test.reshape(len(answer_test), -1)
if index == -1:
#print(len(self.FeatureSet.feature.trainSet_feat[0]),len(answer_train))
self.FeatureSet.feature.trainSet_feat[0] = np.column_stack(
(self.FeatureSet.feature.trainSet_feat[0], answer_train))
self.FeatureSet.feature.testSet_feat[0] = np.column_stack(
(self.FeatureSet.feature.testSet_feat[0], answer_test))
self.FeatureSet.dict[0][toolName] = [
self.FeatureSet.size[0],
self.FeatureSet.size[0] + len(answer_train[0])
]
self.FeatureSet.size[0] += len(answer_train[0])
else:
self.FeatureSet.feature.trainSet_feat[0][:, index[0]:index[1]] = [
i for i in answer_train
]
self.FeatureSet.feature.testSet_feat[0][:, index[0]:index[1]] = [
i for i in answer_test
]
svd_entropy_features_train = []
for i in range(X_train.shape[0]):
##print i
h = svd_entropy(X_train[i, ], 4, 10, W=None)
svd_entropy_features_train.append(h)
svd_entropy_features_test = []
for i in range(X_test.shape[0]):
##print i
h = svd_entropy(X_test[i, ], 4, 10, W=None)
svd_entropy_features_test.append(h)
'''Fisher Information ''' ##Okay
fisher_info_features_train = []
for i in range(X_train.shape[0]):
##print i
h = fisher_info(X_train[i, ], 4, 10, W=None)
fisher_info_features_train.append(h)
fisher_info_features_test = []
for i in range(X_test.shape[0]):
##print i
h = fisher_info(X_test[i, ], 4, 10, W=None)
fisher_info_features_test.append(h)
''' Detrended Fluctuation Analysis''' ###Okay
dfa_features_train = []
for i in range(X_train.shape[0]):
##print i
h = dfa(X_train[i, ], Ave=None, L=None)
dfa_features_train.append(h)
dfa_features_test = []
h = svd_entropy(X_train[i,], 4, 10, W=None)
svd_entropy_features_train.append(h)
svd_entropy_features_test = []
for i in range(X_test.shape[0]):
##print i
h = svd_entropy(X_test[i,], 4, 10, W=None)
svd_entropy_features_test.append(h)
"""Fisher Information """ ##Okay
fisher_info_features_train = []
for i in range(X_train.shape[0]):
##print i
h = fisher_info(X_train[i,], 4, 10, W=None)
fisher_info_features_train.append(h)
fisher_info_features_test = []
for i in range(X_test.shape[0]):
##print i
h = fisher_info(X_test[i,], 4, 10, W=None)
fisher_info_features_test.append(h)
""" Detrended Fluctuation Analysis""" ###Okay
dfa_features_train = []
for i in range(X_train.shape[0]):
##print i
h = dfa(X_train[i,], Ave=None, L=None)
def calculate_features(samples):
data = samples
if not samples:
print("no samples")
return []
band = [0.5, 4, 7, 12, 30]
a = randn(4097)
# approx = pyeeg.ap_entropy(data, 5, 1)
approx = 0
DFA = pyeeg.dfa(data)
first_order_diff = [data[i] - data[i - 1] for i in range(1, len(data))]
fisher_info = pyeeg.fisher_info(data, 1, 1, W=None)
embed_seq = pyeeg.embed_seq(data, 1, 1)
hfd = pyeeg.hfd(data, 6)
hjorth = pyeeg.hjorth(data, D=None)
hurst = pyeeg.hurst(data)
PFD = pyeeg.pfd(data)
sam_ent = pyeeg.samp_entropy(data, 1, 2)
spectral_entropy = pyeeg.spectral_entropy(data,
band,
256,
Power_Ratio=None)
svd = pyeeg.svd_entropy(data, 6, 4, W=None)
PSI = pyeeg.bin_power(data, band, 256)
# # Power Spectral Intensity (PSI) and Relative Intensity Ratio (RIR) Two 1- D v ec t o rs
# # print("bin_power = ", PSI)
# # Petrosian Fractal Dimension (PFD) Ascalar
# print("PFD = ", PFD)
# # Higuchi Fractal Dimension (HFD) Ascalar
# print("hfd = ", hfd)
# # Hjorth mobility and complexity Two s c a la rs
# print("hjorth = ", hjorth)
# # Spectral Entropy (Shannon’s entropy of RIRs) Ascalar
# print("spectral_entropy = ", spectral_entropy)
# # SVD Entropy Ascalar
# print("svd = ", svd)
# # Fisher Information Ascalar
# print("fisher_info = ", fisher_info)
# # Approximate Entropy (ApEn) Ascalar
# print("approx entrophy = ", approx)
# # Detrended Fluctuation Analysis (DFA) Ascalar
# print("DFA = ", DFA)
# # HurstExponent(Hurst) Ascalar
# print("Hurst_Exponent = ", hurst)
# # Build a set of embedding sequences from given time series X with lag Tau and embedding dimension
# print("embed_seq = ", embed_seq)
# # Compute the first order difference of a time series.
# print("first_order_diff = ", first_order_diff)
return {
'approximate': approx,
'DFA': DFA,
'fisher_info': fisher_info,
'embed_seq': embed_seq,
'hfd': hfd,
'hjorth': hjorth,
'hurst': hurst,
'PFD': PFD,
'sam_ent': sam_ent,
'spectral_entropy': spectral_entropy,
'svd': svd,
'PSI': PSI,
'first_order_diff': first_order_diff
}
def compute_pyeeg_feats(self, rec):
# these values are taken from the tuh paper
TAU, DE, Kmax = 4, 10, 5
pwrs, pwrrs, pfds, hfds, mblts, cmplxts, ses, svds, fis, hrsts = [], [], [], [], [], [], [], [], [], []
dfas, apes = [], []
for window_id, window in enumerate(rec.signals):
for window_electrode_id, window_electrode in enumerate(window):
# taken from pyeeg code / paper
electrode_diff = list(np.diff(window_electrode))
M = pyeeg.embed_seq(window_electrode, TAU, DE)
W = scipy.linalg.svd(M, compute_uv=False)
W /= sum(W)
power, power_ratio = self.bin_power(window_electrode,
self.bands,
rec.sampling_freq)
pwrs.extend(list(power))
# mean of power ratio is 1/(len(self.bands)-1)
pwrrs.extend(list(power_ratio))
pfd = pyeeg.pfd(window_electrode, electrode_diff)
pfds.append(pfd)
hfd = pyeeg.hfd(window_electrode, Kmax=Kmax)
hfds.append(hfd)
mobility, complexity = pyeeg.hjorth(window_electrode,
electrode_diff)
mblts.append(mobility)
cmplxts.append(complexity)
se = self.spectral_entropy(window_electrode, self.bands,
rec.sampling_freq, power_ratio)
ses.append(se)
svd = pyeeg.svd_entropy(window_electrode, TAU, DE, W=W)
svds.append(svd)
fi = pyeeg.fisher_info(window_electrode, TAU, DE, W=W)
fis.append(fi)
# this crashes...
# ape = pyeeg.ap_entropy(electrode, M=10, R=0.3*np.std(electrode))
# apes.append(ape)
# takes very very long to compute
# hurst = pyeeg.hurst(electrode)
# hrsts.append(hurst)
# takes very very long to compute
# dfa = pyeeg.dfa(electrode)
# dfas.append(dfa)
pwrs = np.asarray(pwrs).reshape(rec.signals.shape[0],
rec.signals.shape[1],
len(self.bands) - 1)
pwrs = np.mean(pwrs, axis=0)
pwrrs = np.asarray(pwrrs).reshape(rec.signals.shape[0],
rec.signals.shape[1],
len(self.bands) - 1)
pwrrs = np.mean(pwrrs, axis=0)
pfds = np.asarray(pfds).reshape(rec.signals.shape[0],
rec.signals.shape[1])
pfds = np.mean(pfds, axis=0)
hfds = np.asarray(hfds).reshape(rec.signals.shape[0],
rec.signals.shape[1])
hfds = np.mean(hfds, axis=0)
mblts = np.asarray(mblts).reshape(rec.signals.shape[0],
rec.signals.shape[1])
mblts = np.mean(mblts, axis=0)
cmplxts = np.asarray(cmplxts).reshape(rec.signals.shape[0],
rec.signals.shape[1])
cmplxts = np.mean(cmplxts, axis=0)
ses = np.asarray(ses).reshape(rec.signals.shape[0],
rec.signals.shape[1])
ses = np.mean(ses, axis=0)
svds = np.asarray(svds).reshape(rec.signals.shape[0],
rec.signals.shape[1])
svds = np.mean(svds, axis=0)
fis = np.asarray(fis).reshape(rec.signals.shape[0],
rec.signals.shape[1])
fis = np.mean(fis, axis=0)
return list(pwrs.ravel()), list(pwrrs.ravel(
)), pfds, hfds, mblts, cmplxts, ses, svds, fis, apes, hrsts, dfas
"times": 10,
"name": "svd_ent",
"is_original": True,
"fun": lambda x: pyeeg.svd_entropy(x, 2, 3)
},
{
"times": 100,
"name": "svd_ent",
"is_original": False,
"fun": lambda x: univ.svd_entropy(x, 2, 3)
},
{
"times": 10,
"name": "fisher_info",
"is_original": True,
"fun": lambda x: pyeeg.fisher_info(x, 2, 3)
},
{
"times": 100,
"name": "fisher_info",
"is_original": False,
"fun": lambda x: univ.fisher_info(x, 2, 3)
},
{
"times": 100,
"name": "spectral_entropy",
"is_original": True,
"fun": lambda x: pyeeg.spectral_entropy(x, SPECT_ENT_BANDS, 256)
},
{
"times": 100,
def myFeaturesExtractor(
X, myM, myV): # X has to be a matrix where each row is a channel
N = len(X) # number of channels
L = len(X[0])
maxtLyap = min(500, L // 2 + L // 4)
lyapLags = np.arange(maxtLyap) / Fs
# get number of features
nFeatures = nMono * N + N * (N - 1) / 2
# here we initialize the list of features // We will transform it to an array later
featList = np.zeros((int(nFeatures)))
# deal with monovariate features first
for kChan in range(N):
kFeat = 0
mySig = X[kChan, :]
#========== Stats ========================
myMean = myM[kChan]
featList[nMono * kChan + kFeat] = myMean
kFeat += 1
myMax = max(mySig)
featList[nMono * kChan + kFeat] = myMax
kFeat += 1
myMin = min(mySig)
featList[nMono * kChan + kFeat] = myMin
kFeat += 1
peak = max(abs(np.array([myMin, myMax])))
featList[nMono * kChan + kFeat] = peak
kFeat += 1
myVar = myV[kChan]
featList[nMono * kChan + kFeat] = myVar
kFeat += 1
featList[nMono * kChan + kFeat] = sp.skew(mySig)
kFeat += 1
featList[nMono * kChan + kFeat] = sp.kurtosis(mySig)
kFeat += 1
myRMS = rms(mySig)
featList[nMono * kChan + kFeat] = myRMS
kFeat += 1
featList[nMono * kChan + kFeat] = peak / myRMS
kFeat += 1
featList[nMono * kChan + kFeat] = totVar(mySig)
kFeat += 1
featList[nMono * kChan + kFeat] = pyeeg.dfa(mySig)
kFeat += 1
featList[nMono * kChan + kFeat] = pyeeg.hurst(mySig)
kFeat += 1
hMob, hComp = pyeeg.hjorth(mySig)
featList[nMono * kChan + kFeat] = hMob
kFeat += 1
featList[nMono * kChan + kFeat] = hComp
kFeat += 1
## ======== fractal ========================
# Now we need to get the embeding time lag Tau and embeding dmension
ac = delay.acorr(mySig, maxtau=maxTauLag, norm=True, detrend=True)
Tau = firstTrue(ac < corrThresh) # embeding delay
f1 , f2 , f3 = dimension.fnn(mySig, dim=dim, tau=Tau, R=10.0, A=2.0, metric='euclidean',\
window=10,maxnum=None, parallel=True)
myEmDim = firstTrue(f3 < fracThresh)
# Here we construct the Embeding Matrix Em
Em = pyeeg.embed_seq(mySig, Tau, myEmDim)
U, s, Vh = linalg.svd(Em)
W = s / np.sum(s) # list of singular values in decreasing order
FInfo = pyeeg.fisher_info(X, Tau, myEmDim, W=W)
featList[nMono * kChan + kFeat] = FInfo
kFeat += 1
featList[nMono * kChan + kFeat] = Tau
kFeat += 1
featList[nMono * kChan + kFeat] = myEmDim
kFeat += 1
#========================================
PFD = pyeeg.pfd(mySig, D=None)
hfd6 = pyeeg.hfd(mySig, 6)
hfd10 = pyeeg.hfd(mySig, 10)
# Now we fit aline and get its slope to have Lyapunov exponent
divAvg = lyapunov.mle(Em,
maxt=maxtLyap,
window=3 * Tau,
metric='euclidean',
maxnum=None)
poly = np.polyfit(lyapLags,
divAvg,
1,
rcond=None,
full=False,
w=None,
cov=False)
LyapExp = poly[0]
featList[nMono * kChan + kFeat] = PFD
kFeat += 1
featList[nMono * kChan + kFeat] = hfd6
kFeat += 1
featList[nMono * kChan + kFeat] = hfd10
kFeat += 1
featList[nMono * kChan + kFeat] = LyapExp
kFeat += 1
## ======== Entropy ========================
tolerance = 1 / 4
entropyDim = max([myEmDim, PFD])
featList[nMono * kChan + kFeat] = pyeeg.samp_entropy(
mySig, entropyDim, tolerance)
kFeat += 1
featList[nMono * kChan + kFeat] = pyeeg.svd_entropy(mySig,
Tau,
myEmDim,
W=W)
kFeat += 1
# here we compute bin power
power, power_Ratio = pyeeg.bin_power(mySig, freqBins, Fs)
featList[nMono * kChan + kFeat] = pyeeg.spectral_entropy(
mySig, freqBins, Fs, Power_Ratio=power_Ratio)
kFeat += 1
## ======== Spectral ========================
for kBin in range(len(freqBins) - 1):
featList[nMono * kChan + kFeat] = power[kBin]
kFeat += 1
featList[nMono * kChan + kFeat] = power_Ratio[kBin]
kFeat += 1
# deal with multivariate features first
#============ connectivity ==================
corrList = connectome(X)
nConnect = len(corrList)
if N * (N - 1) / 2 != nConnect:
raise ValueError('incorrect number of correlation coeffs')
for kC in range(nConnect):
featList[-nConnect + kC] = corrList[kC]
return featList