def plot_multi_delta_time(ids, **kwargs):
""" Plot delta versus the timestamps
"""
plot = MultiPlot(1, len(ids))
for splot, id in zip(plot.subplots, ids):
ext_timestamps, deltas = get(id)
daystamps = (np.array(ext_timestamps) - min(ext_timestamps)) / 864e11
if max(daystamps) > 3:
idx = next(i for i, v in enumerate(daystamps) if v > 3.2)
deltas = deltas[:idx]
daystamps = daystamps[:idx]
splot.plot(daystamps[::101],
deltas[::101],
mark=None,
linestyle='very thin')
# splot.scatter(daystamps[::400], deltas[::400], mark='*',
# markstyle="mark size=.1pt")
splot.set_axis_options(r'width=%.2f\textwidth' % (.9 / len(ids)))
splot.set_xlimits(0, max(daystamps))
plot.show_xticklabels_for_all()
plot.show_yticklabels(0, 0)
plot.set_xlabel(r'Time in test [days]')
plot.set_ylabel(r'$\Delta t$ [\si{\ns}]')
plot.set_ylimits_for_all(None, -175, 175)
name = 'delta_time/tt_delta_time_' + '_'.join([str(id) for id in ids])
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted delta vs time'
def plot_delta_test(ids, **kwargs):
""" Plot the delta with std
"""
if type(ids) is int:
ids = [ids]
# Define Bins
low = -200
high = 200
bin_size = 1
bins = np.arange(low - .5 * bin_size, high + bin_size, bin_size)
# Begin Figure
plot = Plot()
for id in ids:
ext_timestamps, deltas = get(id)
n, bins = np.histogram(deltas, bins, density=True)
plot.histogram(n, bins)
if kwargs.keys():
plot.set_title('Tijdtest ' + kwargs[kwargs.keys()[0]])
plot.set_xlabel(r'$\Delta$ t (swap - reference) [ns]')
plot.set_ylabel(r'p')
plot.set_xlimits(low, high)
plot.set_ylimits(0., .15)
# Save Figure
if len(ids) == 1:
name = 'tt_delta_hist_%03d' % ids[0]
elif kwargs.keys():
name = 'tt_delta_hist_' + kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted histogram'
def print_delta_results(ids=None):
""" Prints the average delta, the standard deviation and length in days
"""
if not ids:
ids = get_tests(part='id')
if type(ids) is int:
ids = [ids]
for id in ids:
ext_timestamps, deltas = get(id)
print " % 3d %s % 7.2f % 6.2f % 4.2f" % (
id, get_tests(id=id, part='group')[0].ljust(13),
round(np.average(deltas), 2), round(np.std(deltas), 2),
(max(ext_timestamps) - min(ext_timestamps)) / 864e11)
def plot_delta_histogram(ids, **kwargs):
""" Plot a histogram of the deltas
"""
if type(ids) is int:
ids = [ids]
# Bin width
bin_size = 1 # 2.5*n
# Begin Figure
plot = Plot()
for id in ids:
ext_timestamps, deltas = get(id)
low = floor(min(deltas))
high = ceil(max(deltas))
bins = np.arange(low - .5 * bin_size, high + bin_size, bin_size)
n, bins = np.histogram(deltas, bins)
bin_centers = (bins[:-1] + bins[1:]) / 2
popt, pcov = curve_fit(gauss,
bin_centers,
n,
p0=[.15, np.mean(deltas),
np.std(deltas)])
plot.histogram(n, bins)
plot.plot(bin_centers,
gauss(bin_centers, *popt),
mark=None,
linestyle='gray')
if kwargs.keys():
plot.set_title('Tijdtest ' + kwargs[kwargs.keys()[0]])
plot.set_label(r'$\mu={1:.1f}$, $\sigma={2:.1f}$'.format(*popt))
plot.set_xlabel(r'Time difference [ns]')
plot.set_ylabel(r'Counts')
plot.set_xlimits(low, high)
plot.set_ylimits(min=0.)
# Save Figure
if len(ids) == 1:
name = 'delta_histogram/tt_delta_hist_%03d' % ids[0]
elif kwargs.keys():
name = 'delta_histogram/tt_delta_hist_' + kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted histogram'
def print_pulse_frequency(ids=None):
""" Prints the average delta, the standard deviation and length in days
"""
if not ids:
ids = get_tests(part='id')
if type(ids) is int:
ids = [ids]
print " id his/gps/trg mean std max min"
for id in ids:
ts, deltas = get(id)
intervals = np.array(ts[1:]) - np.array(ts[:-1])
print " % 3d %s % 12d % 12d % 12d % 12d" % (
id, get_tests(id=id, part='group')[0].ljust(13),
round(np.average(intervals)), round(
np.std(intervals)), max(intervals), min(intervals))
def plot_multi_box(ids1, ids2, **kwargs):
"""Box Plot like results"""
li = len(ids1) + len(ids2)
# Begin Figure
plot = MultiPlot(1, 2)
for splot, ids in zip(plot.subplots, [ids1, ids2]):
data = [
determine_stats([i for i in get(id)[1] if abs(i) < 100])
for id in ids
]
low, high, avg, std = zip(*data)
x = range(len(ids))
splot.plot(x,
avg,
yerr=std,
mark='o',
markstyle="mark size=1pt",
linestyle=None)
splot.scatter(x, low, mark='x', markstyle="mark size=.75pt")
splot.scatter(x, high, mark='x', markstyle="mark size=.75pt")
splot.set_axis_options(r'height=.67\textwidth, width=%.2f\textwidth' %
((len(ids) * .67 / li)))
splot.set_xlimits(-0.5, len(ids) - 0.5)
splot.set_xticks(x)
plot.set_xlabel(r'Tests')
plot.set_ylabel(r'$\Delta t$ [\si{\ns}]')
plot.set_ylimits_for_all(None, -70, 70)
plot.show_yticklabels(0, 0)
# Save Figure
if len(ids) == 1:
name = 'box/tt_offset_%03d' % ids[0]
elif kwargs.keys():
name = 'box/tt_offset_' + kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted offsets'
def plot_delta_time(ids, **kwargs):
""" Plot delta versus the timestamps
"""
if type(ids) is int:
ids = [ids]
# Begin Figure
plot = Plot()
for index, id in enumerate(ids):
ext_timestamps, deltas = get(id)
# daystamps = (np.array(ext_timestamps) - min(ext_timestamps)) / 864e11
daystamps = (np.array(ext_timestamps) -
min(ext_timestamps)) / 864e11 * 24 * 60
end = min(len(daystamps), 6000)
skip = 3
plot.plot(daystamps[:end:skip],
deltas[:end:skip],
mark=None,
linestyle='very thin')
if kwargs.keys():
plot.set_title('Tijdtest delta time' + kwargs[kwargs.keys()[0]])
# plot.set_xlabel(r'Time in test [days]')
plot.set_xlabel(r'Time in test [minutes]')
plot.set_ylabel(r'$\Delta$ t (swap - reference) [\si{\ns}]')
plot.set_xlimits(0, daystamps[:end][-1])
plot.set_ylimits(-50, 50)
# plot.set_ylimits(-175, 175)
plot.set_axis_options('line join=round')
# Save Figure
if len(ids) == 1:
name = 'delta_time/tt_delta_time_%03d' % ids[0]
elif kwargs.keys():
name = 'delta_time/tt_delta_time_' + kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted delta vs time'
def plot_ns_histogram(ids, **kwargs):
""" Plot a histogram of the nanosecond part of the ext_timestamps
"""
if type(ids) is int:
ids = [ids]
# Define Bins
low = 0
high = int(1e9)
bin_size = 5e6 # 1e7 = 10ms, 1e6 = 1 ms, 1e3 = 1 us
bins = np.arange(low, high + bin_size, bin_size)
# Begin Figure
plot = Plot()
for id in ids:
ext_timestamps, deltas = get(id)
nanoseconds = nanoseconds_from_ext_timestamp(ext_timestamps)
n, bins = np.histogram(nanoseconds, bins, normed=1)
plot.histogram(n, bins)
if kwargs.keys():
plot.set_title('Tijdtest ' + kwargs[kwargs.keys()[0]])
plot.set_xlabel(r'nanosecond part of timestamp [ns]')
plot.set_ylabel(r'p')
plot.set_xlimits(0, 1e9)
plot.set_ylimits(.95e-9, 1.05e-9)
# Save Figure
if len(ids) == 1:
name = 'nanoseconds_histogram/tt_nanos_hist_%03d' % ids[0]
elif kwargs.keys():
name = 'nanoseconds_histogram/tt_nanos_hist_' + kwargs[kwargs.keys()
[0]]
try:
plot.save_as_pdf(PLOT_PATH + name)
except:
print 'tt_analyse: Failed ns hist for %s' % str(ids)
print 'tt_analyse: Plotted histogram'
def plot_box(ids, **kwargs):
"""Box Plot like results"""
if type(ids) is int:
ids = [ids]
x = range(len(ids))
# Begin Figure
plot = Plot()
data = [
determine_stats([i for i in get(id)[1] if abs(i) < 100]) for id in ids
]
low, high, avg, std = zip(*data)
plot.plot(x,
avg,
yerr=std,
mark='o',
markstyle="mark size=1pt",
linestyle=None)
plot.scatter(x, low, mark='x', markstyle="mark size=.5pt")
plot.scatter(x, high, mark='x', markstyle="mark size=.5pt")
# if kwargs.keys():
# plot.set_title('Tijdtest offsets ' + kwargs[kwargs.keys()[0]])
plot.set_xlabel(r'ids')
plot.set_ylabel(r'$\Delta$ t (swap - reference) [\si{\nano\second}]')
plot.set_ylimits(-80, 80)
# Save Figure
if len(ids) == 1:
name = 'box/tt_offset_%03d' % ids[0]
elif kwargs.keys():
name = 'box/tt_offset_' + kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted offsets'
def plot_offset_distribution(ids, **kwargs):
"""Offset distribution"""
# Begin Figure
plot = Plot()
offsets = [
np.average([x for x in get(id)[1] if abs(x) < 100]) for id in ids
]
bins = np.arange(-70, 70, 2)
n, bins = np.histogram(offsets, bins)
plot.histogram(n, bins)
bin_centers = (bins[:-1] + bins[1:]) / 2
popt, pcov = curve_fit(gauss,
bin_centers,
n,
p0=[1., np.mean(offsets),
np.std(offsets)])
plot.plot(bin_centers,
gauss(bin_centers, *popt),
mark=None,
linestyle='gray')
if kwargs.keys():
plot.set_title('Tijdtest offset distribution ' +
kwargs[kwargs.keys()[0]])
plot.set_label(r'$\mu={1:.1f}$, $\sigma={2:.1f}$'.format(*popt))
plot.set_xlabel(r'Offset [\si{\nano\second}]')
plot.set_ylabel(r'Counts')
plot.set_ylimits(min=0)
# Save Figure
name = 'box/tt_offset_distribution'
if kwargs.keys():
name += kwargs[kwargs.keys()[0]]
plot.save_as_pdf(PLOT_PATH + name)
print 'tt_analyse: Plotted offsets'
