def funcNbofAttr(P, G):
endDir = 'G_%.3f_P_%.5f.npy' %(G,P)
try:
pattsx = data2array(dir_prix + '/patterns_' + endDir, mmap_mode="r+")
pattsA = data2array(dir_priA + '/patterns_' + endDir, mmap_mode="r+")
return len(pattsx)
except:
try:
pattsx = data2array(dir_prix + '/allPatt_' + endDir)
pattsA = data2array(dir_priA + '/allPatt_' + endDir)
except:
N = data2array(dco).shape[0]
pattsx = zeros((mmax,N))
pattsA = zeros((mmax,N))
conn = {'connAd': dco,
'normType': '1'}
noise = {'colors': None}
model = {'model': 'HopfieldBasedStatic',
'threshold': 'local',
'tauT': 0,
'P': P,
'G': G}
out = []
other = {'init': 'rand',
'dens': rand(),} #p2,!!!!!!!!!!!!!! RAND
for d in range(mmax):
eva = main.evaCure(evaCon=conn, evaNoi=noise, evaMod=model, out=out, **other)
eva.toEquilibrium()
pattsx[d] = eva.evaMod.x.copy()
pattsA[d] = eva.evaMod.A.copy()
array2data(pattsx, dir_prix + '/allPatt_' + endDir)
array2data(pattsA, dir_priA + '/allPatt_' + endDir)
patts = pattsx # !!!!!!!!!
S = sortBy(patts.mean(1) - patts.mean(), inverse=1)[0]
C1, freq = preClustering(patts[S], sim_coef=sim_coef, sim_func=similarity_Euclidean)
C2, freq = preClustering(patts[S][C1], freq=freq, sim_coef=sim_coef, sim_func=fPearsonCorrelation)
SC, freq = sortBy(freq, inverse=1)
array2data(pattsx[S][C1][C2][SC], dir_prix + '/patterns_' + endDir)
array2data(pattsA[S][C1][C2][SC], dir_priA + '/patterns_' + endDir)
array2data(freq, dir_priT + '/tendances_' + endDir)
os.system('rm ' + dir_prix + '/allPatt_' + endDir)
os.system('rm ' + dir_priA + '/allPatt_' + endDir)
return len(pattsx[S][C1][C2][SC])
def save(_, dir_=None):
'''Save the database in the directory .dir by default or specified during the initialization.
If dir_ is specified, it change the directory of the database while saving it.
'''
try:
size = humanSize(path.getsize(_.dir))
if dir_:
_.dir = dir_
array2data(_, _.dir)
print 'Saved here : %s (%s, %i items)' %(_.dir, size, _.len())
except:
print 'Problem with the adresses.'
return None
mkdir(dir2X)
sim_coef = 0.9
#endFName = '_P_%.2f_f0_%.2f.npy' %(p1,p2)
endFName = '_P_%.2f_G_%.2f.npy' % (p1, p2)
fileName = 'patterns' + endFName
############################################################################################'
# **Main**
if not adressExists(dir2X + fileName):
tic()
pattsX = data2array(dir1X + fileName,
mmap_mode="r+") #.reshape((3300,998))
pattsS = data2array(dir1S + fileName,
mmap_mode="r+") #.reshape((3300,998))
patts = pattsX #!!!
S = sortBy(patts.mean(1) - patts.mean(), inverse=1)[0]
C1, freq = preClustering(patts[S],
sim_coef=sim_coef,
sim_func=similarity_Euclidean)
C2, freq = preClustering(patts[S][C1],
freq=freq,
sim_coef=sim_coef,
sim_func=fPearsonCorrelation)
SC, freq = sortBy(freq, inverse=1)
array2data(pattsX[S][C1][C2][SC], dir2X + fileName)
array2data(pattsS[S][C1][C2][SC], dir2S + fileName)
array2data(freq, dir2X + '/tendances' + endFName)
tac()
eva = main.evaCure(evaCon=conn,
evaNoi=noise,
evaMod=model,
out=out,
**other)
#Run
for i in range(T0):
eva.update()
eva.evaNoi.updateNoise(stdD_x=sx, stdD_T=sT)
for i in range(TF):
eva.update()
#Save
x.append(array(eva.out['x']))
A.append(array(eva.out['A']).mean(1))
del eva
print iii
tac()
############################################################################################'
# **Save**
array2data(array(x), './DL/x_%i.npy' % p1)
del x
array2data(array(A), './DL/A_%i.npy' % p1)
del A
else:
print 'File exists.'
# **Parameters**
while len(sys.argv) > 1:
option = sys.argv[1]; del sys.argv[1]
if option == '-G': G = float(sys.argv[1].replace(',','.')); del sys.argv[1]
elif option == '-sx': sx = float(sys.argv[1].replace(',','.')); del sys.argv[1]
elif option == '-sT': sT = float(sys.argv[1].replace(',','.')); del sys.argv[1]
elif option == '-dir': dir_= str(sys.argv[1]); del sys.argv[1]
else:
print 'Options invalides :',option,'->',sys.argv[0]
mdir = '/home/mathieuG/TCS/%s' %dir_
name = 'G_%.2f_sx_%.3f_sT_%.3f.npy' %(G, sx, sT)
dir_DFC = mdir + '/DFC_998/'
dir_TC = mdir + '/TC_998/'
try:
mkdir(dir_DFC)
except:
pass
############################################################################################'
# **Main**
if not adressExists(dir_DFC + name):
TC = data2array(dir_TC + name)
FCs = windowedFCs(TC, window=6000, jump=100)
DFC = windowedCorrelations(FCs)
array2data(DFC, dir_DFC + name)
############################################################################################'
# **Loop**
tic()
for iii in range(100):
# Init
other.update({'dens':rand()})
eva = main.evaCure(evaCon=conn, evaNoi=noise, evaMod=model, out=out, **other)
#Run
for i in range(T0):
eva.update()
#Save
x.append( array(eva.out['x']) )
A.append( array(eva.out['A']).mean(1) )
del eva
print iii
tac()
############################################################################################'
# **Save**
array2data(array(x), './SL_0/x_%i.npy'%p1); del x
array2data(array(A), './SL_0/A_%i.npy'%p1); del A
else:
print 'File exists.'
noise = {'colors': None}
model = {
'model': 'HopfieldBasedStatic',
'threshold': 'global',
'tauT': 0,
'P': p1,
'G': p2
} #900}
out = []
other = {
'init': 'rand',
'dens': 0.3, #rand(), #p2,
'rperiod': 100
}
for d in range(3300):
eva = main.evaCure(evaCon=conn,
evaNoi=noise,
evaMod=model,
out=out,
**other)
eva.toEquilibrium()
all_pattx[d] = eva.evaMod.x.copy()
all_pattA[d] = eva.evaMod.A.copy()
array2data(all_pattx, dir_prix + dir_sub)
array2data(all_pattA, dir_priA + dir_sub)
eva = main.evaCure(evaCon=conn,
evaNoi=noise,
evaMod=model,
out=out,
**other)
#Run
for i in range(T0):
eva.update()
eva.evaNoi.updateNoise(stdD_x=sx, stdD_T=sT)
for i in range(TF):
eva.update()
#Save
x.append(array(eva.out['x']))
A.append(array(eva.out['A']).mean(1))
del eva
print iii
tac()
############################################################################################'
# **Save**
array2data(array(x), './SG/x_%i.npy' % p1)
del x
array2data(array(A), './SG/A_%i.npy' % p1)
del A
else:
print 'File exists.'
S = sortBy(patts.mean(1) - patts.mean(), inverse=1)[0]
C1, freq = preClustering(patts[S], sim_coef=sim_coef, sim_func=similarity_Euclidean)
C2, freq = preClustering(patts[S][C1], freq=freq, sim_coef=sim_coef, sim_func=fPearsonCorrelation)
SC, freq = sortBy(freq, inverse=1)
array2data(pattsx[S][C1][C2][SC], dir_prix + '/patterns_' + endDir)
array2data(pattsA[S][C1][C2][SC], dir_priA + '/patterns_' + endDir)
array2data(freq, dir_priT + '/tendances_' + endDir)
os.system('rm ' + dir_prix + '/allPatt_' + endDir)
os.system('rm ' + dir_priA + '/allPatt_' + endDir)
return len(pattsx[S][C1][C2][SC])
############################################################################################'
# **Main**
dir_sub = '/dP_G_%.3f_r_%i.npy' %(G, revert)
if not adressExists(dir_priC + dir_sub):
mkdir(dir_prix)
mkdir(dir_priA)
mkdir(dir_priT)
mkdir(dir_priC)
if revert: x = [0.95, 2.00]
else: x = [0.00, 1.05]
dx = paramExplo(funcNbofAttr, nb=[7,1], ax='x', x=x, y=[G], revert=revert)
array2data(dx, dir_priC + dir_sub)
'tauT': 80,
'P': P,
'G': 900.,
}
noise = {'stdD_x': sx, 'stdD_T': sT, 'colors': ['white', 'white']}
out = ['x']
other = {'init': 'rand', 'dens': 0.5, 'rperiod': 100, 'dur': '20m'}
eva = main.evaCure(evaCon=conn,
evaNoi=noise,
evaMod=model,
out=out,
**other)
eva.updateTillEnd()
TC = eva.out['x']
array2data(TC, dir_TC + '/TC_998_' + name)
if not adressExists(dir_FC + '/FC_998_' + name):
try:
TC = array(TC)
except:
TC = data2array(dir_TC + '/TC_998_' + name)
tic()
FC = fPearsonCorrelation(TC)
tac('h')
array2data(FC, dir_FC + '/FC_998_' + name)
'tauT': 80,
'P': P,
'G': 900.,
}
noise = {'stdD_x': sx, 'stdD_T': sT, 'colors': ['white', 'white']}
out = ['x']
other = {'init': 'rand', 'dens': 0.5, 'rperiod': 100, 'dur': '20m'}
eva = main.evaCure(evaCon=conn,
evaNoi=noise,
evaMod=model,
out=out,
**other)
eva.updateTillEnd()
TC = eva.out['x']
array2data(TC[len(TC) / 20:], dir_TC + '/TC_998_' + name)
if not adressExists(dir_FC + '/FC_998_' + name):
try:
TC = array(TC)
except:
TC = data2array(dir_TC + '/TC_998_' + name)
tic()
FC = fPearsonCorrelation(TC)
tac('h')
array2data(FC, dir_FC + '/FC_998_' + name)
