#gamma = 1.0/1.547
gamma = 1.0/10.547
p_over_q = 1.01
A = 1.0
deltaM = 0.4
deltaG = 0.0
charges = [[+1,+1], [-1,-1], [+1,-1], [-1,+1]]
maxes = []
for i in range(0,4):
q1 = charges[i][0]
q2 = charges[i][1]
N = pdfs.pdf_bmixing(deltat,[gamma,p_over_q,deltaM,deltaG,q1,q2])
maxes.append(max(N))
subplots[0][i].plot(deltat,N,'-',markersize=2)
max_prob = max(maxes)
#'''
events = [np.array([]),np.array([]),np.array([]),np.array([])]
n=0
nevents = 1000
print "Generating %d events." % (nevents)
while n<nevents:
if n%1000==0:
def main():
############################################################################
# Make a figure on which to plot stuff.
# This would be the same as making a TCanvas object.
############################################################################
figs = []
subplots = []
for i in xrange(2):
figs.append(plt.figure(figsize=(8,6),dpi=100,facecolor='w',edgecolor='k'))
subplots.append([])
for j in range(0,4):
subplots[i].append(figs[i].add_subplot(2,2,j+1))
############################################################################
# Generate values drawn from a normal (Gaussian) distribution.
############################################################################
deltat_min = -20
deltat_max = 20
deltat_range = deltat_max-deltat_min
deltat_mc = []
for i in range(0,4):
deltat_mc.append(deltat_range*np.random.rand(10000) + deltat_min)
print "deltat MC: %d" % (len(deltat_mc))
deltat = np.linspace(deltat_min,deltat_max,1000)
gamma = 1.0/1.547
p_over_q = 1.01
A = 1.0
deltaM = 0.4
deltaG = 0.0
charges = [[+1,+1], [-1,-1], [+1,-1], [-1,+1]]
maxes = []
for i in range(0,4):
q1 = charges[i][0]
q2 = charges[i][1]
N = pdfs.pdf_bmixing(deltat,[gamma,p_over_q,deltaM,deltaG,q1,q2])
maxes.append(max(N))
subplots[0][i].plot(deltat,N,'-',markersize=2)
max_prob = max(maxes)
#'''
events = [np.array([]),np.array([]),np.array([]),np.array([])]
n=0
nevents = 1000
print "Generating %d events." % (nevents)
while n<nevents:
if n%1000==0:
print n
for i in range(0,4):
q1 = charges[i][0]
q2 = charges[i][1]
val = deltat_range*np.random.rand()+deltat_min
prob = pdfs.pdf_bmixing(val,[gamma,p_over_q,deltaM,deltaG,q1,q2])
test = max_prob*np.random.rand()
if test<prob:
events[i] = np.append(events[i],val)
n += 1
for i in range(0,4):
#subplots[1][i].hist(events[i],bins=50)
figs[1].add_subplot(2,2,i+1)
lch.hist_err(events[i],bins=50)
subplots[1][i].set_xlim(deltat_min,deltat_max)
subplots[1][i].set_ylim(0)
#subplots[1][i].set_ylim(0,nevents/16)
Npp = len(events[0])
Nmm = len(events[1])
Npm = len(events[2])
Nmp = len(events[3])
print "%d %d %d %d" % (Npp,Nmm,Npm,Nmp)
Acp = (Npp-Nmm)/float(Npp+Nmm)
deltaAcp = np.sqrt((sqrt(Npp)/Npp)**2 + (sqrt(Nmm)/Nmm)**2)
print "Acp: %f +/- %f" % (Acp,deltaAcp)
Acp = 2*(1-np.abs(1.0/p_over_q))
print "Acp: %f" % (Acp)
#'''
# Fit function.
gamma = 1.0/1.547
p_over_q = 1.01
A = 1.0
deltaM = 0.4
deltaG = 0.0
#N = pdfs.pdf_bmixing(deltat,[gamma,p_over_q,deltaM,deltaG,q1,q2])
n0 = Npp
n1 = Nmm
n2 = Npm
n3 = Nmp
p0 = [gamma,p_over_q,deltaM,deltaG,n0,n1,n2,n3]
print p0
#p1 = sp.optimize.fmin(pdfs.extended_maximum_likelihood_function,p0,args=(events,deltat_mc), maxiter=10000, maxfun=10000)
#print p1
data = [events,deltat_mc]
m = minuit.Minuit(pdfs.extended_maximum_likelihood_function_minuit,p=p0)
print m.values
m.migrad()
# Need this command to display the figure.
plt.show()
def main():
############################################################################
# Make a figure on which to plot stuff.
# This would be the same as making a TCanvas object.
############################################################################
figs = []
subplots = []
for i in xrange(1):
figs.append(plt.figure(figsize=(8,6),dpi=100,facecolor='w',edgecolor='k'))
subplots.append([])
for j in range(0,4):
subplots[i].append(figs[i].add_subplot(2,2,j+1))
############################################################################
# Generate values drawn from a normal (Gaussian) distribution.
############################################################################
deltat_min = -20
deltat_max = 20
deltat_range = deltat_max-deltat_min
deltat_mc = []
for i in range(0,4):
deltat_mc.append(deltat_range*np.random.rand(10000) + deltat_min)
print "deltat MC: %d" % (len(deltat_mc))
deltat = np.linspace(deltat_min,deltat_max,1000)
gamma = 1.0/1.547
p_over_q = 1.01
A = 1.0
deltaM = 0.4
deltaG = 0.0
charges = [[+1,+1], [-1,-1], [+1,-1], [-1,+1]]
maxes = []
for i in range(0,4):
q1 = charges[i][0]
q2 = charges[i][1]
N = pdfs.pdf_bmixing(deltat,[gamma,p_over_q,deltaM,deltaG,q1,q2])
maxes.append(max(N))
subplots[0][i].plot(deltat,N,'-',markersize=2)
max_prob = max(maxes)
'''
events = [np.array([]),np.array([]),np.array([]),np.array([])]
n=0
nevents = 100
print "Generating %d events." % (nevents)
while n<nevents:
if n%1000==0:
print n
for i in range(0,4):
q1 = charges[i][0]
q2 = charges[i][1]
val = deltat_range*np.random.rand()+deltat_min
prob = pdfs.pdf_bmixing(val,[gamma,p_over_q,deltaM,deltaG,q1,q2])
test = max_prob*np.random.rand()
if test<prob:
events[i] = np.append(events[i],val)
n += 1
for i in range(0,4):
#subplots[1][i].hist(events[i],bins=50)
figs[1].add_subplot(2,2,i+1)
lch.hist_err(events[i],bins=50)
subplots[1][i].set_xlim(deltat_min,deltat_max)
subplots[1][i].set_ylim(0)
#subplots[1][i].set_ylim(0,nevents/16)
Npp = len(events[0])
Nmm = len(events[1])
Npm = len(events[2])
Nmp = len(events[3])
print "%d %d %d %d" % (Npp,Nmm,Npm,Nmp)
Acp = (Npp-Nmm)/float(Npp+Nmm)
deltaAcp = np.sqrt((sqrt(Npp)/Npp)**2 + (sqrt(Nmm)/Nmm)**2)
print "Acp: %f +/- %f" % (Acp,deltaAcp)
Acp = 2*(1-np.abs(1.0/p_over_q))
print "Acp: %f" % (Acp)
'''
'''
# Fit function.
gamma = 1.0/1.547
p_over_q = 1.01
A = 1.0
deltaM = 0.4
deltaG = 0.0
#N = pdfs.pdf_bmixing(deltat,[gamma,p_over_q,deltaM,deltaG,q1,q2])
n0 = Npp
n1 = Nmm
n2 = Npm
n3 = Nmp
p0 = [gamma,p_over_q,deltaM,deltaG,n0,n1,n2,n3]
print p0
#p1 = sp.optimize.fmin(pdfs.extended_maximum_likelihood_function,p0,args=(events,deltat_mc), maxiter=10000, maxfun=10000)
#print p1
#data = [events,deltat_mc]
#m = minuit.Minuit(pdfs.extended_maximum_likelihood_function_minuit,p=p0)
#print m.values
#m.migrad()
'''
# Need this command to display the figure.
plt.show()
