def _squeezing(fname, coupling, ens):
with open(get_path(__file__, 'single_well_squeezing_wigner_' + ens + '_Na200.pickle')) as f:
wigner = pickle.load(f)
suffix = '_c' if coupling else '_nc'
interaction = (100.4, 80.8, 95.5) if coupling else (100.4, 0., 100.4)
tau_exact = numpy.linspace(1e-3, 120 if coupling else 20, 200)
s_exact = get_exact_S(tau_exact, 200, *interaction)
tau_wigner = wigner['tau' + suffix]
s_wigner = wigner['s_pi2' + suffix]
s_wigner_err = wigner['s_pi2' + suffix + '_err']
s_exact = numpy.log10(s_exact) * 10
tau_wigner, s_wigner_bot, s_wigner_top = mplh.crop_bounds(
tau_wigner,
numpy.log10(s_wigner - s_wigner_err) * 10,
numpy.log10(s_wigner + s_wigner_err) * 10,
(0, (120 if coupling else 20), -15, 1))
fig = mplh.figure(width=0.5)
s = fig.add_subplot(111,
xlabel='$\\tau$',
ylabel='$S_{\\theta + \\pi/2}$ (dB)')
s.plot(tau_exact, s_exact, color="black")
s.fill_between(
tau_wigner, s_wigner_bot, s_wigner_top,
facecolor=mplh.color.f.blue.lightest,
linewidth=0)
s.text(
18 if coupling else 3,
-1,
"interaction on" if coupling else "interaction off")
s.set_xlim((0, 120 if coupling else 20))
s.set_ylim((-15, 1))
s.set_aspect((5 ** 0.5 - 1) / 2 * mplh.aspect_modifier(s))
fig.text(0.01, 0.92, '(b)' if coupling else '(a)', fontweight='bold')
fig.tight_layout(pad=0.3)
fig.savefig(fname)
def _squeezing_err(fname, coupling):
suffix = '_c' if coupling else '_nc'
colors = {
'10k': mplh.color.f.blue,
'1k': mplh.color.f.red,
'100': mplh.color.f.green}
dashes = {
'10k': '-',
'1k': '--'}
fig = mplh.figure(width=0.5)
s = fig.add_subplot(111,
xlabel='$\\tau$',
ylabel='Relative errors')
for tr in ('1k', '10k'):
with open(get_path(__file__, 'single_well_squeezing_wigner_' + tr + '_Na200.pickle')) as f:
wigner = pickle.load(f)
tau_wigner = wigner['tau' + suffix]
s_wigner = wigner['s_pi2' + suffix]
s_wigner_err = wigner['s_pi2' + suffix + '_err']
interaction = (100.4, 80.8, 95.5) if coupling else (100.4, 0., 100.4)
s_exact = get_exact_S(tau_wigner, 200, *interaction)
s_exact[0] = 1.
diff = numpy.abs(s_wigner - s_exact) / s_exact
err = s_wigner_err / s_exact
tau_diff, min_diff, max_diff = mplh.crop_bounds(
tau_wigner, diff-err, diff+err, (0, (120 if coupling else 20), 0, 0.1))
s.fill_between(tau_diff, min_diff, max_diff,
facecolor=colors[tr].light,
linewidth=0,
alpha=0.5)
s.plot(tau_wigner, diff, color=colors[tr].dark, dashes=mplh.dash[dashes[tr]])
s.text(
72 if coupling else 12,
0.085,
"interaction on" if coupling else "interaction off")
if coupling:
s.text(55, 0.06, '$20,000$')
s.text(55, 0.052, 'trajectories')
s.text(19, 0.02, '$200,000$')
s.text(19, 0.012, 'trajectories')
else:
s.text(13, 0.06, '$20,000$')
s.text(13, 0.052, 'trajectories')
s.text(4, 0.02, '$200,000$')
s.text(4, 0.012, 'trajectories')
s.set_xlim((0, 120 if coupling else 20))
s.set_ylim((0, 0.1))
s.set_aspect((5 ** 0.5 - 1) / 2 * mplh.aspect_modifier(s))
fig.text(0.01, 0.92, '(b)' if coupling else '(a)', fontweight='bold')
fig.tight_layout(pad=0.3)
fig.savefig(fname)
def _squeezing_N_err(fname, coupling):
suffix = '_c' if coupling else '_nc'
fig = mplh.figure(width=0.5)
s = fig.add_subplot(111,
xlabel='$\\tau / \\tau_c(N)$',
ylabel='Relative errors')
colors = {
20: mplh.color.f.blue,
200: mplh.color.f.red,
2000: mplh.color.f.green}
dashes = {
20: '-',
200: '--',
2000: ':'}
for Na in (20, 200, 2000):
with open(get_path(__file__, 'single_well_squeezing_wigner_10k_Na' + str(Na) + '.pickle')) as f:
wigner = pickle.load(f)
tau_wigner = wigner['tau' + suffix]
s_wigner = wigner['s_pi2' + suffix]
s_wigner_err = wigner['s_pi2' + suffix + '_err']
interaction = (100.4, 80.8, 95.5) if coupling else (100.4, 0., 100.4)
s_exact = get_exact_S(tau_wigner, Na, *interaction)
s_exact[0] = 1.
tau = tau_wigner / tau_wigner[-1]
diff = numpy.abs(s_wigner - s_exact) / s_exact
err = s_wigner_err / s_exact
tau_diff, min_diff, max_diff = mplh.crop_bounds(tau, diff-err, diff+err, (0, 1., 0, 0.16))
s.fill_between(tau_diff, min_diff, max_diff,
facecolor=colors[Na].light,
linewidth=0,
alpha=0.5)
s.plot(tau, diff, color=colors[Na].dark, dashes=mplh.dash[dashes[Na]])
s.text(
0.6,
0.135,
"interaction on" if coupling else "interaction off")
if coupling:
s.text(0.12, 0.12, '$N=20$')
s.text(0.25, 0.062, '$N=200$')
s.text(0.17, 0.03, '$N=2000$')
else:
s.text(0.1, 0.12, '$N=20$')
s.text(0.27, 0.082, '$N=200$')
s.text(0.16, 0.01, '$N=2000$')
s.set_xlim((0, 1.))
s.set_ylim((0, 0.16))
s.set_aspect((5 ** 0.5 - 1) / 2 * mplh.aspect_modifier(s))
fig.text(0.01, 0.92, '(b)' if coupling else '(a)', fontweight='bold')
fig.tight_layout(pad=0.3)
fig.savefig(fname)
def _feshbach_squeezing(fname, losses):
with open(get_path(__file__, 'feshbach_squeezing' + ('' if losses else '_no_losses') + '.json')) as f:
sq = json.load(f)
datasets = [
('80.0', mplh.color.f.blue, '-', ),
('85.0', mplh.color.f.red, '--'),
('90.0', mplh.color.f.green, ':'),
('95.0', mplh.color.f.yellow, '-.'),
]
t_sq = numpy.array(sq['times'])
fig = mplh.figure(width=0.5)
subplot = fig.add_subplot(111)
for a12, color, linestyle in datasets:
xi2_sq = numpy.array(sq['xi2_' + a12])
xi2_sq_err = numpy.array(sq['xi2_' + a12 + '_err'])
xi2_log = 10 * numpy.log10(xi2_sq)
err_down = 10 * numpy.log10(xi2_sq - xi2_sq_err)
err_up = 10 * numpy.log10(xi2_sq + xi2_sq_err)
t_err, err_down, err_up = mplh.crop_bounds(
t_sq, err_down, err_up, (0, 0.1, (-13 if losses else -20), 1))
positive = err_up > err_down
#subplot.fill_between(
# t_err * 1e3, err_down, err_up,
# facecolor=color.lightest,
#interpolate=True,
# linewidth=0)
subplot.plot(t_sq * 1e3, xi2_log, color=color.main, dashes=mplh.dash[linestyle])
subplot.plot([0, 100], [0, 0], color='grey', linewidth=0.5,
linestyle='-.', dashes=mplh.dash['-.'])
subplot.set_xlim(xmin=0, xmax=100)
subplot.set_ylim(ymin=-13 if losses else -20 , ymax=1)
subplot.set_xlabel('$T$ (ms)')
subplot.set_ylabel('$\\xi^2$ (dB)')
subplot.text(75,
1 - 2.5 if losses else 1 - (2.5 / 14. * 21.),
'1-2 losses' if losses else 'no losses')
if losses:
subplot.text(40, -2, '$80.0\,r_B$')
subplot.text(70, -8.5, '$85.0\,r_B$')
subplot.text(40, -11.0, '$90.0\,r_B$')
subplot.text(43, -6.75, '$95.0\,r_B$')
else:
subplot.text(35, -16.5, '$80.0\,r_B$')
subplot.text(70, -15., '$85.0\,r_B$')
subplot.text(75, -12., '$90.0\,r_B$')
subplot.text(75, -8., '$95.0\,r_B$')
fig.text(0.01, 0.92, '(b)' if losses else '(a)', fontweight='bold')
fig.tight_layout(pad=0.3)
fig.savefig(fname)