def combinedTest(N, a12, gamma12, losses):
t = None
ii = None
nn1 = None
nn2 = None
for i in xrange(N):
times, i, n1, n2 = testUncertainties(a12, gamma12, losses)
if ii is None:
ii = i
nn1 = n1
nn2 = n2
t = times
else:
for j in xrange(len(ii)):
ii[j] = numpy.concatenate([ii[j], i[j]])
nn1[j] = numpy.concatenate([nn1[j], n1[j]])
nn2[j] = numpy.concatenate([nn2[j], n2[j]])
xi = []
if losses:
m = Model(N=44000, detuning=-37,
nvx=8, nvy=8, nvz=64, dt_evo=4e-5, ensembles=1024 * N, e_cut=1e6,
fx=97.0, fy=97.0 * 1.03, fz=11.69,
gamma12=gamma12, gamma22=7.7e-20,
a12=a12, a22=95.4)
else:
m = Model(N=44000, detuning=-37,
nvx=8, nvy=8, nvz=64, dt_evo=4e-5, ensembles=1024 * N, e_cut=1e6,
fx=97.0, fy=97.0 * 1.03, fz=11.69,
gamma111=0, gamma12=0, gamma22=0,
a12=a12, a22=95.4)
constants = Constants(m, double=True)
env = envs.cuda()
u = Uncertainty(env, constants)
for j in xrange(len(ii)):
xi.append(u._getXiSquared(ii[j], nn1[j], nn2[j]))
env.release()
return XYData("a12 = " + str(a12) + ("" if losses else ", no losses"),
times * 1e3, 10 * numpy.log10(numpy.array(xi)),
xname="T (ms)", yname="$\\xi^2$ (dB)")
def combinedTest(N, a12, gamma12, losses):
t = None
ii = None
nn1 = None
nn2 = None
for i in xrange(N):
times, i, n1, n2 = testUncertainties(a12, gamma12, losses)
if ii is None:
ii = i
nn1 = n1
nn2 = n2
t = times
else:
for j in xrange(len(ii)):
ii[j] = numpy.concatenate([ii[j], i[j]])
nn1[j] = numpy.concatenate([nn1[j], n1[j]])
nn2[j] = numpy.concatenate([nn2[j], n2[j]])
xi = []
if losses:
m = Model(N=44000,
detuning=-37,
nvx=8,
nvy=8,
nvz=64,
dt_evo=4e-5,
ensembles=1024 * N,
e_cut=1e6,
fx=97.0,
fy=97.0 * 1.03,
fz=11.69,
gamma12=gamma12,
gamma22=7.7e-20,
a12=a12,
a22=95.4)
else:
m = Model(N=44000,
detuning=-37,
nvx=8,
nvy=8,
nvz=64,
dt_evo=4e-5,
ensembles=1024 * N,
e_cut=1e6,
fx=97.0,
fy=97.0 * 1.03,
fz=11.69,
gamma111=0,
gamma12=0,
gamma22=0,
a12=a12,
a22=95.4)
constants = Constants(m, double=True)
env = envs.cuda()
u = Uncertainty(env, constants)
for j in xrange(len(ii)):
xi.append(u._getXiSquared(ii[j], nn1[j], nn2[j]))
env.release()
return XYData("a12 = " + str(a12) + ("" if losses else ", no losses"),
times * 1e3,
10 * numpy.log10(numpy.array(xi)),
xname="T (ms)",
yname="$\\xi^2$ (dB)")
def testRephasing(wigner):
if wigner:
nvx = nvy = 8
nvz = 64
dt_evo = 4e-5
else:
nvx = nvy = 16
nvz = 128
dt_evo = 1e-5
m = Model(N=55000, detuning=-37,
nvx=nvx, nvy=nvy, nvz=nvz, dt_evo=dt_evo, ensembles=64, e_cut=1e6,
fx=97.0, fy=97.0 * 1.03, fz=11.69,
gamma12=1.53e-20, gamma22=7.7e-20,
a12=97.99, a22=95.57)
t_max = 8.0
t_step = 0.05
constants = Constants(m, double=True)
env = envs.cuda(device_num=0)
evolution = SplitStepEvolution(env, constants)
pulse = Pulse(env, constants)
stats = ParticleStatistics(env, constants)
unc = Uncertainty(env, constants)
gs = GPEGroundState(env, constants)
times = [0.0]
vis = [1.0]
phnoise = []
phis = []
ypss = []
cloud = gs.createCloud()
if wigner:
cloud.toWigner()
phnoise.append(stats.getPhaseNoise(cloud.a, cloud.b))
phi, yps = unc.getSpins(cloud.a, cloud.b)
phis.append(phi)
ypss.append(yps)
pulse.apply(cloud, 0.5 * math.pi, matrix=True)
if wigner:
name = 'rephasing_wigner'
else:
name = 'rephasing_gpe'
t = t_step
while t <= t_max:
evolution.run(cloud, t_step / 2, noise=wigner)
print "Pre-pi step finished, t=" + str(cloud.time)
new_cloud = cloud.copy()
pulse.apply(new_cloud, math.pi, matrix=True)
evolution.run(new_cloud, t / 2, noise=wigner)
print "Post-pi step finished"
times.append(t)
vis.append(stats.getVisibility(new_cloud.a, new_cloud.b))
print "Visibility=" + str(vis[-1])
if wigner:
phnoise.append(stats.getPhaseNoise(new_cloud.a, new_cloud.b))
phi, yps = unc.getSpins(new_cloud.a, new_cloud.b)
phis.append(phi)
ypss.append(yps)
XYData(name, numpy.array(times), numpy.array(vis),
ymin=0, ymax=1, xname="T, s", yname="$\\mathcal{V}$").save(name + '_vis.json')
if wigner:
XYData(name, numpy.array(times), numpy.array(phnoise), ymin=0,
xname="T, s", yname="Phase noise, rad").save(name + '_phnoise.json')
Data('spins',
['name', 'phi', 'yps', 'time', 'timename', 'phiname', 'ypsname'],
time=numpy.array(times), phi=numpy.array(phis), yps=numpy.array(ypss), timename="T (s)",
phiname="Spin, azimuth, rad", ypsname="Spin, inclination, rad",
name=name).save(name + "_phnoise_points.pickle")
XYPlot([XYData.load(name + '_vis.json')]).save(name + '_vis.pdf')
del new_cloud
t += t_step
env.release()
def testRephasing(wigner):
if wigner:
nvx = nvy = 8
nvz = 64
dt_evo = 4e-5
else:
nvx = nvy = 16
nvz = 128
dt_evo = 1e-5
m = Model(N=55000,
detuning=-37,
nvx=nvx,
nvy=nvy,
nvz=nvz,
dt_evo=dt_evo,
ensembles=64,
e_cut=1e6,
fx=97.0,
fy=97.0 * 1.03,
fz=11.69,
gamma12=1.53e-20,
gamma22=7.7e-20,
a12=97.99,
a22=95.57)
t_max = 8.0
t_step = 0.05
constants = Constants(m, double=True)
env = envs.cuda(device_num=0)
evolution = SplitStepEvolution(env, constants)
pulse = Pulse(env, constants)
stats = ParticleStatistics(env, constants)
unc = Uncertainty(env, constants)
gs = GPEGroundState(env, constants)
times = [0.0]
vis = [1.0]
phnoise = []
phis = []
ypss = []
cloud = gs.createCloud()
if wigner:
cloud.toWigner()
phnoise.append(stats.getPhaseNoise(cloud.a, cloud.b))
phi, yps = unc.getSpins(cloud.a, cloud.b)
phis.append(phi)
ypss.append(yps)
pulse.apply(cloud, 0.5 * math.pi, matrix=True)
if wigner:
name = 'rephasing_wigner'
else:
name = 'rephasing_gpe'
t = t_step
while t <= t_max:
evolution.run(cloud, t_step / 2, noise=wigner)
print "Pre-pi step finished, t=" + str(cloud.time)
new_cloud = cloud.copy()
pulse.apply(new_cloud, math.pi, matrix=True)
evolution.run(new_cloud, t / 2, noise=wigner)
print "Post-pi step finished"
times.append(t)
vis.append(stats.getVisibility(new_cloud.a, new_cloud.b))
print "Visibility=" + str(vis[-1])
if wigner:
phnoise.append(stats.getPhaseNoise(new_cloud.a, new_cloud.b))
phi, yps = unc.getSpins(new_cloud.a, new_cloud.b)
phis.append(phi)
ypss.append(yps)
XYData(name,
numpy.array(times),
numpy.array(vis),
ymin=0,
ymax=1,
xname="T, s",
yname="$\\mathcal{V}$").save(name + '_vis.json')
if wigner:
XYData(name,
numpy.array(times),
numpy.array(phnoise),
ymin=0,
xname="T, s",
yname="Phase noise, rad").save(name + '_phnoise.json')
Data('spins', [
'name', 'phi', 'yps', 'time', 'timename', 'phiname', 'ypsname'
],
time=numpy.array(times),
phi=numpy.array(phis),
yps=numpy.array(ypss),
timename="T (s)",
phiname="Spin, azimuth, rad",
ypsname="Spin, inclination, rad",
name=name).save(name + "_phnoise_points.pickle")
XYPlot([XYData.load(name + '_vis.json')]).save(name + '_vis.pdf')
del new_cloud
t += t_step
env.release()
