X = np.array(X)
y = np.array(y)
X = X.astype(float)
y = y.astype(float)
nTRs = len(y)
corrC = np.corrcoef(ALL.T)
corrC = np.abs(
corrC) # DO ABS here because negative correlatoion is fine!
thr = tm.bc_alpha(corrC, alpha=0.05)
#corrC[corrC < thr] = 0
signif = []
for i, e in enumerate(corrC[-1]):
if tm.RtoP(e, nTRs) < thr:
signif.append(i)
signif = signif[:-1]
X = X.T[signif]
X = X.T
# CHANGE THIS TO SK LEARN!
#OLS STUFF
#X = sm.add_constant(X)
#rez = sm.OLS(y, X).fit()
#B = rez.params
import sklearn.linear_model
reg = sklearn.linear_model.Lasso(alpha=0.1)
reg.fit(X, y)
fCount = np.zeros(30) for l in inFile: ALL.append(l) ALL = np.array(ALL) nTRs = len(ALL) ALL = ALL.astype(float) corrC = np.corrcoef(ALL.T) corrC = np.abs(corrC) # DO ABS here because negative correlatoion is fine! #thr = 0.05 thr = tm.bc_alpha(corrC, alpha=0.05) #thr = tm.FDR_thresh_corrmat(corrC, nTRs, alpha=0.05) #print thr #corrC[tm.RtoP(corrC,nTRs) < thr] = 0 DATA = corrC[:, -1] DATA = [tm.RtoP(x, nTRs) for x in DATA] #print DATA #DATA[DATA > thr] = 100 #print DATA for i in range(len(DATA)): if DATA[i] < thr: if i == (int(lasts[lastsCount]) - 1): fCount[29] += 1 elif i == 29: fCount[int(lasts[lastsCount]) - 1] += 1 else: fCount[i] += 1 fCount[int(lasts[lastsCount]) - 1] = 1