-
Notifications
You must be signed in to change notification settings - Fork 0
/
utilityFunctions.py
470 lines (389 loc) · 19 KB
/
utilityFunctions.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
import numpy as np
import scipy as sp
import pandas as pd
from scipy import signal, arange, fft, fromstring, roll
from scipy.signal import butter, lfilter, ricker
import os
import glob
from sklearn.neighbors import KNeighborsClassifier
from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.feature_selection import RFE
from sklearn.svm import SVR
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
from sklearn import metrics
from sklearn.cluster import DBSCAN
from sklearn.metrics import recall_score, precision_score, f1_score, accuracy_score
from scipy.stats import stats
from sklearn.decomposition import PCA
def fastClassOutputs(N,X,y,featureNumber):
clf = QuadraticDiscriminantAnalysis()
ca1finalAcc,ca1finalF1,finalFeatures,finalLength,ca1As=dualClass(N,clf,X,y,featureNumber)
#cb1fsAcc,cb1fsF1,finalFeatures,finalLength,cb1As=fsClass(N,clf,X,y,featureNumber)
clf = GaussianNB()
#ca2finalAcc,ca2finalF1,finalFeatures,finalLength,ca2As=dualClass(N,clf,X,y,featureNumber)
#cb2fsAcc,cb2fsF1,finalFeatures,finalLength,cb2As=fsClass(N,clf,X,y,featureNumber)
clf = SVC(gamma=2, C=1)
ca3finalAcc,ca3finalF1,finalFeatures,finalLength,ca3As=dualClass(N,clf,X,y,featureNumber)
#cb3fsAcc,cb3fsF1,finalFeatures,finalLength,cb3As=fsClass(N,clf,X,y,featureNumber)
clf = KNeighborsClassifier(n_neighbors=3)
ca4finalAcc,ca4finalF1,finalFeatures,finalLength,ca4As=dualClass(N,clf,X,y,featureNumber)
#cb4fsAcc,cb4fsF1,finalFeatures,finalLength,cb4As=fsClass(N,clf,X,y,featureNumber)
return(ca1finalAcc,ca1finalF1,ca1As,ca3finalAcc,ca3finalF1,ca3As,ca4finalAcc,ca4finalF1,ca4As)
def classOldOutputs(N,X,y,featureNumber):
clf = QuadraticDiscriminantAnalysis()
ca1finalAcc,ca1finalF1,finalFeatures,finalLength,ca1As=dualClass(N,clf,X,y,featureNumber)
cb1fsAcc,cb1fsF1,finalFeatures,finalLength,cb1As=fsClass(N,clf,X,y,featureNumber)
clf = GaussianNB()
ca2finalAcc,ca2finalF1,finalFeatures,finalLength,ca2As=dualClass(N,clf,X,y,featureNumber)
cb2fsAcc,cb2fsF1,finalFeatures,finalLength,cb2As=fsClass(N,clf,X,y,featureNumber)
clf = SVC(gamma=2, C=1)
ca3finalAcc,ca3finalF1,finalFeatures,finalLength,ca3As=dualClass(N,clf,X,y,featureNumber)
cb3fsAcc,cb3fsF1,finalFeatures,finalLength,cb3As=fsClass(N,clf,X,y,featureNumber)
clf = KNeighborsClassifier(n_neighbors=3)
ca4finalAcc,ca4finalF1,finalFeatures,finalLength,ca4As=dualClass(N,clf,X,y,featureNumber)
cb4fsAcc,cb4fsF1,finalFeatures,finalLength,cb4As=fsClass(N,clf,X,y,featureNumber)
clf = LogisticRegression(random_state=0)
ca5finalAcc,ca5finalF1,finalFeatures,finalLength,ca5As=dualClass(N,clf,X,y,featureNumber)
cb5fsAcc,cb5fsF1,finalFeatures,finalLength,cb5As=fsClass(N,clf,X,y,featureNumber)
clf = RandomForestClassifier(max_depth=2, random_state=0)
ca6finalAcc,ca6finalF1,finalFeatures,finalLength,ca6As=dualClass(N,clf,X,y,featureNumber)
cb6fsAcc,cb6fsF1,finalFeatures,finalLength,cb6As=fsClass(N,clf,X,y,featureNumber)
return(ca1finalAcc,ca1finalF1,cb1fsAcc,cb1fsF1,ca2finalAcc,ca2finalF1,cb2fsAcc,cb2fsF1,ca3finalAcc,ca3finalF1,cb3fsAcc,cb3fsF1,ca4finalAcc,ca4finalF1,cb4fsAcc,cb4fsF1,ca1As,ca2As,ca3As,ca4As,cb1As,cb2As,cb3As,cb4As)
def classOutputs(N,X,y,featureNumber):
clf = QuadraticDiscriminantAnalysis()
ca1finalAcc,ca1finalF1,finalFeatures,finalLength,ca1As=dualClass(N,clf,X,y,featureNumber)
cb1fsAcc,cb1fsF1,finalFeatures,finalLength,cb1As=fsClass(N,clf,X,y,featureNumber)
clf = GaussianNB()
ca2finalAcc,ca2finalF1,finalFeatures,finalLength,ca2As=dualClass(N,clf,X,y,featureNumber)
cb2fsAcc,cb2fsF1,finalFeatures,finalLength,cb2As=fsClass(N,clf,X,y,featureNumber)
clf = SVC(gamma=2, C=1)
ca3finalAcc,ca3finalF1,finalFeatures,finalLength,ca3As=dualClass(N,clf,X,y,featureNumber)
cb3fsAcc,cb3fsF1,finalFeatures,finalLength,cb3As=fsClass(N,clf,X,y,featureNumber)
clf = KNeighborsClassifier(n_neighbors=3)
ca4finalAcc,ca4finalF1,finalFeatures,finalLength,ca4As=dualClass(N,clf,X,y,featureNumber)
cb4fsAcc,cb4fsF1,finalFeatures,finalLength,cb4As=fsClass(N,clf,X,y,featureNumber)
clf = LogisticRegression(random_state=0)
ca5finalAcc,ca5finalF1,finalFeatures,finalLength,ca5As=dualClass(N,clf,X,y,featureNumber)
cb5fsAcc,cb5fsF1,finalFeatures,finalLength,cb5As=fsClass(N,clf,X,y,featureNumber)
clf = RandomForestClassifier(max_depth=2, random_state=0)
ca6finalAcc,ca6finalF1,finalFeatures,finalLength,ca6As=dualClass(N,clf,X,y,featureNumber)
cb6fsAcc,cb6fsF1,finalFeatures,finalLength,cb6As=fsClass(N,clf,X,y,featureNumber)
return(ca1finalAcc,ca1finalF1,cb1fsAcc,cb1fsF1,ca2finalAcc,ca2finalF1,cb2fsAcc,cb2fsF1,ca3finalAcc,ca3finalF1,cb3fsAcc,cb3fsF1,ca4finalAcc,ca4finalF1,cb4fsAcc,cb4fsF1,ca5finalAcc,ca5finalF1,cb5fsAcc,cb5fsF1,ca6finalAcc,ca6finalF1,cb6fsAcc,cb6fsF1)
def featureSelect(X, y, featureNumber, catToSearch):
locks1 = np.where(y==catToSearch)
locks2 = np.where(y!=catToSearch)
X1 = np.squeeze(X[locks1,:])
X2 = np.squeeze(X[locks2,:])
totalLength=np.array([np.shape(X1)[0],np.shape(X2)[0]])
topFeatures=eegFeatureReducer(X1, X2, featureNumber)
aX1=balancedMatrix(X1, totalLength)
aX2=balancedMatrix(X2, totalLength)
atopFeatures=eegFeatureReducer(aX1, aX2, featureNumber)
aFeatures=np.unique(np.vstack([topFeatures, atopFeatures])).flatten()
return (aFeatures,totalLength,X1,X2)
def speedClass(X1, X2):
Xa = np.vstack([X1, X2])
t0 = 1*np.ones([1, len(X1)])
t1 = 0*np.ones([1, len(X2)])
targets = np.hstack([t0, t1])
ya = np.transpose(np.ravel(targets))
return (Xa,ya)
def dirClass(N,clf,Xa,ya):
scores = cross_val_score(clf, Xa, ya, cv=N)
avScore=scores.mean()
avStd=scores.std()
scores = cross_val_score(clf, Xa, ya, cv=N, scoring='f1_macro')
f1Score=scores.mean()
f1Std=scores.std()
scores = cross_val_score(clf, Xa, ya, cv=N, scoring='roc_auc')
aucScore=scores.mean()
aucStd=scores.std()
return (avScore,f1Score,avStd,f1Std,aucScore,aucStd)
#from numpy import mean,cov,double,cumsum,dot,linalg,array,rank
#from pylab import plot,subplot,axis,stem,show,figure
def princomp(A):
M = (A-mean(A.T,axis=1)).T # subtract the mean (along columns)
[latent,coeff] = linalg.eig(cov(M)) # attention:not always sorted
score = dot(coeff.T,M) # projection of the data in the new space
return coeff,score,latent
def dualClass(N,clf,X,y,featureNumber):
nuNu=list()
runCats=np.squeeze(np.unique(y))
nuInd=list()
nuAv=list()
nuF1=list()
nuAs=list()
# features1=np.array([339 340 341 342 160 161 181 182 183 329 336 337 338 718 719 720 184 185 186 203 204 205 190 191 192 193 194 168 169 170 310 164 165 166 217 218 219 168 169 170 404 405 406 189 190 191 203 204 205 155 156 157 409 410 411 201 202 203 343 344 345 165 166 167 303 304 305 188 189 190 195 196 197 161 162 163 342 343 344 100 181 182 239 240 870 165 166 167 293 294 295 325 326 327 328 329 330 168 169 170 287 288 289 168 169 170 133 134 135 168 169 658 659 660 153 154 155 378 379 380 63 64 65 203 204 205 413 414 415 145 146 147 168 169 170 168 169 170 332 333 334 190 191 192 298 299 300 186 202 203 204 344 345 191 192 193 203 204 205 238 239 240 288 289 290 236 237 238 239 240 168 169 170 205 153 154 155 293 294 295 162 163 164 165 166 131 132 133 134 164 165 166 168 169 170 168 169 170 709 778 779 168 169 170 232 233 234 378 379 380 605 606 607 98 99 100 181 182 183 9 10 11 308 309 310 98 99 100])
f2=np.array([9, 10, 11, 63, 64, 65, 98, 99, 100, 131, 132, 133, 134, 135, 145, 146, 147, 153, 154, 155, 156, 157, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 181, 182, 183, 184, 185, 186, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 201, 202, 203, 204, 205, 217, 218, 219, 232, 233, 234, 236, 237, 238, 239, 240, 287, 288, 289, 290, 293, 294, 295, 298, 299, 300, 303, 304, 305, 308, 309, 310, 325, 326, 327, 328, 329, 330, 332, 333, 334, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 378, 379, 380, 404, 405, 406, 409, 410, 411, 413, 414, 415, 605, 606, 607, 658, 659, 660, 709, 718, 719, 720, 778, 779, 870])
for ii in runCats:
aFeatures,totalLength,X1,X2=featureSelect(X, y, featureNumber, ii)
Xa,ya=speedClass(X1, X2)
#Xa = np.squeeze(Xa[:,f2])
#pcaData = PCA(n_components=3)
#Xa = pcaData.fit_transform(Xa)
#princomp(X)
avScore,f1Score,avStd,f1Std,aucScore,aucStd=dirClass(N,clf,Xa,ya)
nuNu.append(aFeatures)
nuInd.append(totalLength)
nuAv.append(avScore)
nuF1.append(f1Score)
nuAs.append(aucScore)
finalFeatures=np.asarray(nuNu,dtype=object)
finalLength=np.asarray(totalLength)
finalAcc=np.mean(np.asarray(nuAv))
finalF1=np.mean(np.asarray(nuF1))
finalAs=np.mean(np.asarray(nuAs))
return(finalAcc,finalF1,finalFeatures,finalLength,finalAs)
def fsClass(N,clf,X,y,featureNumber):
nuNu=list()
runCats=np.squeeze(np.unique(y))
nuInd=list()
nuAv=list()
nuF1=list()
nuAs=list()
for ii in runCats:
aFeatures,totalLength,X1,X2=featureSelect(X, y, featureNumber, ii)
Xa,ya=speedClass(X1, X2)
Xa=np.squeeze(Xa[:,aFeatures])
avScore,f1Score,avStd,f1Std,aucScore,aucStd=dirClass(N,clf,Xa,ya)
nuNu.append(aFeatures)
nuInd.append(totalLength)
nuAv.append(avScore)
nuF1.append(f1Score)
nuAs.append(aucScore)
finalFeatures=np.asarray(nuNu,dtype=object)
finalLength=np.asarray(totalLength)
finalAcc=np.mean(np.asarray(nuAv))
finalF1=np.mean(np.asarray(nuF1))
finalAs=np.mean(np.asarray(nuAs))
return(finalAcc,finalF1,finalFeatures,finalLength,finalAs)
def pairLoader(subName):
outNamData=subName+'_Data.csv'
outNamLabels=subName+'_Labels.csv'
X = np.genfromtxt(outNamData, delimiter=',')
y = np.genfromtxt(outNamLabels, delimiter=',')
X = X.astype(float)
y = y.astype(int)
X[np.isnan(X)] = 0
X[np.isinf(X)] = 0
y[np.isnan(y)] = 0
y[np.isinf(y)] = 0
lims=np.shape(X)
x0=int(lims[0])
x1=int(lims[1])
X=np.squeeze(X[1:x0,1:x1])
runCats=np.squeeze(np.unique(y))
return(X,y)
def ghostFeatures(rawData, indVal, chanNum, fs, lowcut, highcut, pcti, windows):
i1=np.squeeze(indVal[0])
can1=int(chanNum)
ses1=np.squeeze(rawData[i1[0]:i1[1],:])
singChan=ses1[0::,can1]
w1=[0,fs]
w2=[fs,np.min([(2*fs-1),len(singChan)])]
f1a = featureExtraction(singChan[int(w1[0]):int(w1[1])], fs, lowcut, highcut, pcti)
f1b = featureExtraction(singChan[int(w2[0]):int(w2[1])], fs, lowcut, highcut, pcti)
#f1=featureCreation(singChan, fs, lowcut, highcut, pcti, windows)
f1=np.concatenate((f1a,f1b),axis=0)
return(f1)
def ghostHeap(rawData, indVal, fs, lowcut, highcut, pcti, windows):
# fa=ghostFeatures(rawData, indVal, 0, fs, lowcut, highcut, pcti, windows)
fb=ghostFeatures(rawData, indVal, 1, fs, lowcut, highcut, pcti, windows)
fc=ghostFeatures(rawData, indVal, 2, fs, lowcut, highcut, pcti, windows)
fd=ghostFeatures(rawData, indVal, 3, fs, lowcut, highcut, pcti, windows)
fe=ghostFeatures(rawData, indVal, 4, fs, lowcut, highcut, pcti, windows)
ff=ghostFeatures(rawData, indVal, 5, fs, lowcut, highcut, pcti, windows)
fg=ghostFeatures(rawData, indVal, 6, fs, lowcut, highcut, pcti, windows)
fh=ghostFeatures(rawData, indVal, 7, fs, lowcut, highcut, pcti, windows)
fi=ghostFeatures(rawData, indVal, 8, fs, lowcut, highcut, pcti, windows)
fj=ghostFeatures(rawData, indVal, 9, fs, lowcut, highcut, pcti, windows)
fk=ghostFeatures(rawData, indVal, 10, fs, lowcut, highcut, pcti, windows)
fl=ghostFeatures(rawData, indVal, 11, fs, lowcut, highcut, pcti, windows)
fm=ghostFeatures(rawData, indVal, 12, fs, lowcut, highcut, pcti, windows)
fn=ghostFeatures(rawData, indVal, 13, fs, lowcut, highcut, pcti, windows)
fo=ghostFeatures(rawData, indVal, 14, fs, lowcut, highcut, pcti, windows)
fp=ghostFeatures(rawData, indVal, 15, fs, lowcut, highcut, pcti, windows)
fq=ghostFeatures(rawData, indVal, 16, fs, lowcut, highcut, pcti, windows)
#fa1=np.concatenate((fa,fb,fc,fd,fe,ff,fg,fh,fi,fj,fk,fl,fm,fn,fo,fp,fq),axis=0)
fa1=np.concatenate((fb,fc,fd,fe,ff,fg,fh,fi,fj,fk,fl,fm,fn,fo,fp,fq),axis=0)
return(fa1)
def ghostVector(rawData, fs, lowcut, highcut, pcti, h1, h2, h3, h4, h5, windows):
f1=ghostHeap(rawData, h1, fs, lowcut, highcut, pcti, windows)
f2=ghostHeap(rawData, h2, fs, lowcut, highcut, pcti, windows)
f3=ghostHeap(rawData, h3, fs, lowcut, highcut, pcti, windows)
f4=ghostHeap(rawData, h4, fs, lowcut, highcut, pcti, windows)
f5=ghostHeap(rawData, h5, fs, lowcut, highcut, pcti, windows)
featureVector=np.concatenate((f1,f2,f3,f4,f5),axis=0)
return(featureVector)
def eegFeatureExtraction(df, fs, lowcut, highcut, pcti):
chan1 = df.iloc[:, 2]
chan2 = df.iloc[:, 3]
chan3 = df.iloc[:, 4]
chan4 = df.iloc[:, 5]
# rotating the vectors to array
c1 = np.real(np.asarray(chan1))
c2 = np.real(np.asarray(chan2))
c3 = np.real(np.asarray(chan3))
c4 = np.real(np.asarray(chan4))
# Normalizing these arrays
c1 = c1-np.mean(c1)
c2 = c2-np.mean(c2)
c3 = c1-np.mean(c3)
c4 = c4-np.mean(c4)
c1 = c1[fs::]
f1 = featureExtraction(c1, fs, lowcut, highcut, pcti)
features = np.squeeze(np.shape(f1))
c2 = c2[fs::]
c3 = c3[fs::]
c4 = c4[fs::]
lengthFile = np.floor(np.squeeze(np.shape(c1))/np.float(4*fs))
lbnds = np.arange(0, (lengthFile-1))
ubnds = np.arange(1, (lengthFile))
capper = np.min([len(lbnds), len(ubnds)])
lbnds = 4*fs*lbnds[0:capper]
ubnds = 4*fs*ubnds[0:capper]
featureMatrix = np.zeros((capper, (4*features)))
for ix in range(0, capper):
s1 = featureExtraction(
c1[int(lbnds[ix]):int(ubnds[ix])], fs, lowcut, highcut, pcti)
s2 = featureExtraction(
c2[int(lbnds[ix]):int(ubnds[ix])], fs, lowcut, highcut, pcti)
s3 = featureExtraction(
c3[int(lbnds[ix]):int(ubnds[ix])], fs, lowcut, highcut, pcti)
s4 = featureExtraction(
c4[int(lbnds[ix]):int(ubnds[ix])], fs, lowcut, highcut, pcti)
featall = np.concatenate((s1, s2, s3, s4), axis=0)
featall = np.squeeze(featall[0:(4*features)])
featureMatrix[int(ix), :] = featall
return (featureMatrix)
def butter_bandpass(lowcut, highcut, fs, order=4):
nyq = 0.5 * fs
low = (lowcut / nyq)
high = (highcut / nyq)
if high >= 1:
high = .99
if low <= 0:
low = .001
b, a = butter(order, [low, high], btype='band')
return b, a
def butter_bandpass_filter(data, lowcut, highcut, fs, order=4):
b, a = butter_bandpass(lowcut, highcut, fs, order=order)
y = lfilter(b, a, data)
return y
def welchProc(data, fs):
# wsize=round(fs/10)
f, P = signal.welch(data, fs)
return f, P
def peakFinder(f, P):
peakFLoc = np.where(P == np.amax(P))
peakFLoc = peakFLoc[0]
peakF = f[peakFLoc]
vrms = np.sqrt(P.max())
return peakF, peakFLoc, vrms
def smooth(x, window_len=11, window='hanning'):
s = np.r_[x[window_len-1:0:-1], x, x[-2:-window_len-1:-1]]
if window == 'flat': # moving average
w = np.ones(window_len, 'd')
else:
w = eval('np.'+window+'(window_len)')
y = np.convolve(w/w.sum(), s, mode='valid')
return y
def featureCreation(data, fs, lowcut, highcut, pcti, windows):
feature_vector = np.array([])
seconds_per_window = np.floor(len(data)/fs) / windows
window_length = np.floor(seconds_per_window * fs)
j=0
while j < len(data):
win_low = int(j)
win_high = int(j+window_length)
if win_high < len(data):
print(str(win_low) + ' ' + str(win_high))
current_data = data[win_low:win_high]
current_feature = featureExtraction(current_data, fs, lowcut, highcut, pcti)
feature_vector = np.hstack((feature_vector, current_feature))
j = j + window_length
#for j in range(np.floor(len(data)/window_length)):
# window_low = ((j-1) / windows) * len(data)
# window_high = (j / windows) * len(data)
# current_data = data[:,i]
# current_feature = featureExtraction(current_data, fs, lowcut, highcut, pcti)
# feature_vector = np.hstack((feature_vector, current_feature))
return feature_vector
def featureExtraction(data, fs, lowcut, highcut, pcti):
widths = np.arange(1, 31)
data[np.isnan(data)] = 0
data[np.isinf(data)] = 0
intensityPcti = np.percentile(data, pcti)
dMean = np.mean(data)
data = data-np.mean(data)
data = smooth(data.flatten())
data = butter_bandpass_filter(data, lowcut, highcut, fs, order=4)
data = signal.cwt(data, signal.ricker, widths)
f, P = welchProc(data, fs)
peakFLoc = np.where(P == np.amax(P))
peakFLoc = peakFLoc[0]
peakF = f[peakFLoc]
vrms = np.sqrt(P.max())
Psum = np.sum(P, axis=1)
Psum = Psum.flatten()
# print(np.shape(Psum))
# print(np.shape(vrms))
# print(np.shape(peakF))
# print(np.shape(peakFLoc))
# print(np.shape(intensityPcti))
featureVector = np.hstack(
(Psum.flatten(), vrms, peakF, peakFLoc, dMean, intensityPcti))
# print(featureVector)
featureVector[np.isnan(featureVector)] = 0
featureVector[np.isinf(featureVector)] = 0
# print(featureVector)
return featureVector
#def balancedMatrix(a, totalLength):
def balMatrix(a, totalLength):
maxLen = np.max(totalLength)
minLen = np.min(totalLength)
ratioL = np.floor(maxLen/minLen)
finalR = np.ceil(minLen*ratioL)
aT = np.copy(a)
features = np.shape(a)[1]
aTT = np.zeros([1, features])
for ii in range(0, int(ratioL)):
aTT1 = np.copy(aT)
aTT = np.vstack([aTT1, aTT])
aTT = np.squeeze(aTT)
aTT = aTT[0:(maxLen-1), :]
aTT = np.squeeze(aTT)
return (aTT)
def balancedMatrix(a, totalLength):
#def balMatrix(a, totalLength):
minLen = np.min(totalLength)
[xw,xh]=np.shape(a)
orders=np.random.permutation(xw)
truncatedOrd=np.squeeze(orders[0:minLen])
aTT = a[truncatedOrd, :]
aTT = np.squeeze(aTT)
return (aTT)
def eegFeatureReducer(featureMatrixA, featureMatrixB, featureNumber):
m0 = np.mean(featureMatrixA, axis=0)
m1 = np.mean(featureMatrixB, axis=0)
distancesVec = np.abs(m0-m1)
tempR = np.argpartition(-distancesVec, featureNumber)
resultArgs = tempR[:featureNumber]
topFeatures = np.flip(resultArgs)
return (topFeatures)
def featureReducer(Xf, yf, features):
estimator = SVR(kernel="linear")
selector = RFE(estimator, features, step=15)
selector = selector.fit(Xf, yf)
topFeatures = np.where(selector.ranking_ == 1)
Xnew = np.squeeze(Xf[:, topFeatures])
return (Xnew, topFeatures)
def crossValClass(clf, X, y, xfold):
scores = cross_val_score(clf, X, y, cv=xfold)
print("Accuracy: %0.2f (+/- %0.2f)" % (scores.mean(), scores.std() * 2))
scores = cross_val_score(clf, X, y, cv=xfold, scoring='f1_macro')
print("F1 Score: %0.2f (+/- %0.2f)" % (scores.mean(), scores.std() * 2))