def raster(trains, time_unit=pq.ms, show_lines=True, events=None, epochs=None):
""" Create a new plotting window with a rasterplot of spiketrains.
:param dict trains: Dictionary of spike trains indexed by a
Neo object (Unit or Segment).
:param Quantity time_unit: Unit of X-Axis.
:param bool show_lines: Determines if a horizontal line will be shown
for each spike train.
:param sequence events: A sequence of neo `Event` objects that will
be marked on the plot.
"""
if not trains:
raise SpykeException('No spike trains for rasterplot')
if not time_unit:
time_unit = pq.ms
win_title = 'Spike Trains'
win = PlotDialog(toolbar=True, wintitle=win_title, major_grid=False)
pW = BaseCurveWidget(win)
plot = pW.plot
if events is None:
events = []
if epochs is None:
epochs = []
offset = len(trains)
legend_items = []
for u, t in trains.iteritems():
color = helper.get_object_color(u)
train = helper.add_spikes(plot, t, color, 2, 21, offset, u.name,
time_unit)
if u.name:
legend_items.append(train)
if show_lines:
plot.add_item(
make.curve([
t.t_start.rescale(time_unit),
t.t_stop.rescale(time_unit)
], [offset, offset],
color='k'))
offset -= 1
helper.add_epochs(plot, epochs, time_unit)
helper.add_events(plot, events, time_unit)
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM, time_unit.dimensionality.string)
win.add_plot_widget(pW, 0)
legend = make.legend(restrict_items=legend_items)
plot.add_item(legend)
win.add_legend_option([legend], True)
if len(trains) > 1:
plot.set_axis_limits(BasePlot.Y_LEFT, 0.5, len(trains) + 0.5)
win.add_custom_curve_tools()
win.show()
return win
def spikes(spikes,
axes_style,
strong=None,
anti_alias=False,
fade=1.0,
subplot_layout=0,
time_unit=pq.ms,
progress=None):
""" Create a plot dialog with spike waveforms. Assumes that all spikes
have waveforms with the same number of channels.
:param dict spikes: A dictionary of :class:`neo.core.Spike` lists.
:param int axes_style: Plotting mode. The following values are possible:
- 1: Show each channel in a seperate plot, split vertically.
- 2: Show each channel in a separate plot, split horizontally.
- 3: Show each key of ``spikes`` in a separate plot,
channels are split vertically.
- 4: Show each key of ``spikes`` in a separate plot,
channels are split horizontally.
- 5: Show all channels in the same plot, split vertically.
- 6: Show all channels in the same plot, split horizontally.
:param dict strong: A dictionary of :class:`neo.core.Spike` lists. When
given, these spikes are shown as thick lines on top of the regular
spikes in the respective plots.
:param bool anti_alias: Determines whether an antialiased plot is created.
:param float fade: Vary transparency by segment. For values > 0, the first
spike for each unit is displayed with the corresponding alpha
value and alpha is linearly interpolated until it is 1 for the
last spike. For values < 0, alpha is 1 for the first spike and
``fade`` for the last spike. Does not affect spikes from ``strong``.
:param bool subplot_layout: The way subplots are arranged on the window:
- 0: Linear - horizontally or vertically,
depending on ``axis_style``.
- 1: Square - this layout tries to have the same number of plots per
row and per column.
:param Quantity time_unit: Unit of X-Axis.
:param progress: Set this parameter to report progress.
:type progress: :class:`spykeutils.progress_indicator.ProgressIndicator`
"""
if (not spikes or sum((len(l) for l in spikes.itervalues())) < 1) and \
(not strong or sum((len(l) for l in strong.itervalues())) < 1):
raise SpykeException('No spikes for spike waveform plot!')
if not progress:
progress = ProgressIndicator()
if strong is None:
strong = {}
progress.begin('Creating waveform plot')
progress.set_ticks(
sum((len(l) for l in spikes.itervalues())) +
sum((len(l) for l in strong.itervalues())))
win_title = 'Spike waveforms'
win = PlotDialog(toolbar=True, wintitle=win_title)
try:
ref_spike = spikes[spikes.keys()[0]][0]
except IndexError:
ref_spike = strong[strong.keys()[0]][0]
if ref_spike.waveform is None:
raise SpykeException('Cannot create waveform plot: At least one spike '
'has no waveform or sampling rate!')
ref_units = ref_spike.waveform.units
channels = range(ref_spike.waveform.shape[1])
# Keys from spikes and strong without duplicates in original order
seen = set()
indices = [
k for k in spikes.keys() + strong.keys()
if k not in seen and not seen.add(k)
]
if axes_style <= 2: # Separate channel plots
for c in channels:
pw = BaseCurveWidget(win)
plot = pw.plot
plot.set_antialiasing(anti_alias)
for u in spikes:
color = helper.get_object_color(u)
qcol = Qt.QColor(color)
alpha = fade if fade > 0.0 else 1.0
alpha_step = 1.0 - fade if fade > 0.0 else -1.0 - fade
alpha_step /= len(spikes[u])
if len(spikes[u]) == 1:
alpha = 1.0
for s in spikes[u]:
if s.waveform is None or s.sampling_rate is None:
raise SpykeException('Cannot create waveform plot: '
'At least one spike has no '
'waveform or sampling rate!')
x = (sp.arange(s.waveform.shape[0]) /
s.sampling_rate).rescale(time_unit)
curve = make.curve(x,
s.waveform[:, c].rescale(ref_units),
title=u.name,
color=color)
qcol.setAlphaF(alpha)
curve.setPen(Qt.QPen(qcol))
alpha += alpha_step
plot.add_item(curve)
progress.step()
for u in strong:
color = helper.get_object_color(u)
for s in strong[u]:
x = (sp.arange(s.waveform.shape[0]) /
s.sampling_rate).rescale(time_unit)
outline = make.curve(x,
s.waveform[:, c].rescale(ref_units),
color='#000000',
linewidth=4)
curve = make.curve(x,
s.waveform[:, c].rescale(ref_units),
color=color,
linewidth=2)
plot.add_item(outline)
plot.add_item(curve)
progress.step()
_add_plot(plot, pw, win, c, len(channels), subplot_layout,
axes_style, time_unit, ref_units)
helper.make_window_legend(win, indices, True)
elif axes_style > 4: # Only one plot needed
pw = BaseCurveWidget(win)
plot = pw.plot
plot.set_antialiasing(anti_alias)
if axes_style == 6: # Horizontal split
l = _split_plot_hor(channels, spikes, strong, fade, ref_units,
time_unit, progress, plot)
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM,
time_unit.dimensionality.string)
else: # Vertical split
channels.reverse()
max_offset = _find_y_offset(channels, spikes, strong, ref_units)
l = _split_plot_ver(channels, spikes, strong, fade, ref_units,
time_unit, progress, max_offset, plot)
plot.set_axis_title(BasePlot.Y_LEFT, 'Voltage')
plot.set_axis_unit(BasePlot.Y_LEFT,
ref_units.dimensionality.string)
win.add_plot_widget(pw, 0)
win.add_legend_option([l], True)
else: # One plot per unit
if axes_style == 3:
channels.reverse()
max_offset = _find_y_offset(channels, spikes, strong, ref_units)
for i, u in enumerate(indices):
pw = BaseCurveWidget(win)
plot = pw.plot
plot.set_antialiasing(anti_alias)
spk = {}
if u in spikes:
spk[u] = spikes[u]
st = {}
if u in strong:
st[u] = strong[u]
if axes_style == 3: # Vertical split
_split_plot_ver(channels, spk, st, fade, ref_units, time_unit,
progress, max_offset, plot)
else: # Horizontal split
_split_plot_hor(channels, spk, st, fade, ref_units, time_unit,
progress, plot)
_add_plot(plot, pw, win, i, len(indices), subplot_layout,
axes_style, time_unit, ref_units)
win.add_custom_curve_tools()
progress.done()
win.show()
if axes_style <= 2:
if len(channels) > 1:
win.add_x_synchronization_option(True, channels)
win.add_y_synchronization_option(True, channels)
elif axes_style <= 4:
if len(spikes) > 1:
win.add_x_synchronization_option(True, range(len(spikes)))
win.add_y_synchronization_option(True, range(len(spikes)))
return win
def sde(trains,
events=None,
start=0 * pq.ms,
stop=None,
kernel_size=100 * pq.ms,
optimize_steps=0,
minimum_kernel=10 * pq.ms,
maximum_kernel=500 * pq.ms,
kernel=None,
time_unit=pq.ms,
progress=None):
""" Create a spike density estimation plot.
The spike density estimations give an estimate of the instantaneous
rate. Optionally finds optimal kernel size for given data.
:param dict trains: A dictionary of :class:`neo.core.SpikeTrain` lists.
:param dict events: A dictionary (with the same indices as ``trains``)
of Event objects or lists of Event objects. In case of lists,
the first event in the list will be used for alignment. The events
will be at time 0 on the plot. If None, spike trains are used
unmodified.
:param start: The desired time for the start of the first bin. It
will be recalculated if there are spike trains which start later
than this time. This parameter can be negative (which could be
useful when aligning on events).
:type start: Quantity scalar
:param stop: The desired time for the end of the last bin. It will
be recalculated if there are spike trains which end earlier
than this time.
:type stop: Quantity scalar
:param kernel_size: A uniform kernel size for all spike trains.
Only used if optimization of kernel sizes is not used (i.e.
``optimize_steps`` is 0).
:type kernel_size: Quantity scalar
:param int optimize_steps: The number of different kernel sizes tried
between ``minimum_kernel`` and ``maximum_kernel``.
If 0, ``kernel_size`` will be used.
:param minimum_kernel: The minimum kernel size to try in optimization.
:type minimum_kernel: Quantity scalar
:param maximum_kernel: The maximum kernel size to try in optimization.
:type maximum_kernel: Quantity scalar
:param kernel: The kernel function or instance to use, should accept
two parameters: A ndarray of distances and a kernel size.
The total area under the kernel function should be 1.
Automatic optimization assumes a Gaussian kernel and will
likely not produce optimal results for different kernels.
Default: Gaussian kernel
:type kernel: func or :class:`spykeutils.signal_processing.Kernel`
:param Quantity time_unit: Unit of X-Axis.
:param progress: Set this parameter to report progress.
:type progress: :class:`spykeutils.progress_indicator.ProgressIndicator`
"""
if not progress:
progress = ProgressIndicator()
start.units = time_unit
if stop:
stop.units = time_unit
kernel_size.units = time_unit
minimum_kernel.units = time_unit
maximum_kernel.units = time_unit
if kernel is None:
kernel = signal_processing.GaussianKernel(100 * pq.ms)
# Align spike trains
for u in trains:
if events:
trains[u] = rate_estimation.aligned_spike_trains(trains[u], events)
# Calculate spike density estimation
if optimize_steps:
steps = sp.logspace(sp.log10(minimum_kernel), sp.log10(maximum_kernel),
optimize_steps) * time_unit
sde, kernel_size, eval_points = \
rate_estimation.spike_density_estimation(
trains, start, stop,
optimize_steps=steps, kernel=kernel,
progress=progress)
else:
sde, kernel_size, eval_points = \
rate_estimation.spike_density_estimation(
trains, start, stop,
kernel_size=kernel_size, kernel=kernel,
progress=progress)
progress.done()
if not sde:
raise SpykeException('No spike trains for SDE!')
# Plot
win_title = 'Kernel Density Estimation'
win = PlotDialog(toolbar=True, wintitle=win_title)
pW = BaseCurveWidget(win)
plot = pW.plot
plot.set_antialiasing(True)
for u in trains:
if u and u.name:
name = u.name
else:
name = 'Unknown'
curve = make.curve(
eval_points,
sde[u],
title='%s, Kernel width %.2f %s' %
(name, kernel_size[u], time_unit.dimensionality.string),
color=helper.get_object_color(u))
plot.add_item(curve)
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM, eval_points.dimensionality.string)
plot.set_axis_title(BasePlot.Y_LEFT, 'Rate')
plot.set_axis_unit(BasePlot.Y_LEFT, 'Hz')
l = make.legend()
plot.add_item(l)
win.add_plot_widget(pW, 0)
win.add_custom_curve_tools()
win.add_legend_option([l], True)
win.show()
return win
def cross_correlogram(trains,
bin_size,
max_lag=500 * pq.ms,
border_correction=True,
per_second=True,
square=False,
time_unit=pq.ms,
progress=None):
""" Create (cross-)correlograms from a dictionary of spike train
lists for different units.
:param dict trains: Dictionary of :class:`neo.core.SpikeTrain` lists.
:param bin_size: Bin size (time).
:type bin_size: Quantity scalar
:param max_lag: Maximum time lag for which spikes are considered
(end time of calculated correlogram).
:type max_lag: Quantity scalar
:param bool border_correction: Apply correction for less data at higher
timelags.
:param bool per_second: If ``True``, the y-axis is count per second,
otherwise it is count per spike train.
:param bool square: If ``True``, the plot will include all
cross-correlograms, even if they are just mirrored versions of each
other. The autocorrelograms are displayed as the diagonal of a
square plot matrix. If ``False``, mirrored plots are omitted.
:param Quantity time_unit: Unit of X-Axis.
:param progress: Set this parameter to report progress.
:type progress: :class:`spykeutils.progress_indicator.ProgressIndicator`
"""
if not trains:
raise SpykeException('No spike trains for correlogram')
if not progress:
progress = ProgressIndicator()
win_title = 'Correlogram | Bin size ' + str(bin_size)
progress.begin('Creating correlogram')
progress.set_status('Calculating...')
win = PlotDialog(toolbar=True,
wintitle=win_title,
min_plot_width=150,
min_plot_height=100)
correlograms, bins = correlogram(trains, bin_size, max_lag,
border_correction, per_second, time_unit,
progress)
x = bins[:-1] + bin_size / 2
crlgs = []
indices = correlograms.keys()
for i1 in xrange(len(indices)):
start_i = 0
if not square:
start_i = i1
for i2 in xrange(start_i, len(indices)):
crlgs.append((correlograms[indices[i1]][indices[i2]], indices[i1],
indices[i2]))
columns = int(sp.sqrt(len(crlgs)))
legends = []
for i, c in enumerate(crlgs):
legend_items = []
pW = BaseCurveWidget(win)
plot = pW.plot
plot.set_antialiasing(True)
plot.add_item(make.curve(x, c[0]))
# Create legend
color = helper.get_object_color(c[1])
color_curve = make.curve([], [],
c[1].name,
color,
'NoPen',
linewidth=1,
marker='Rect',
markerfacecolor=color,
markeredgecolor=color)
legend_items.append(color_curve)
plot.add_item(color_curve)
if c[1] != c[2]:
color = helper.get_object_color(c[2])
color_curve = make.curve([], [],
c[2].name,
color,
'NoPen',
linewidth=1,
marker='Rect',
markerfacecolor=color,
markeredgecolor=color)
legend_items.append(color_curve)
plot.add_item(color_curve)
legends.append(make.legend(restrict_items=legend_items))
plot.add_item(legends[-1])
if i >= len(crlgs) - columns:
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM,
time_unit.dimensionality.string)
if i % columns == 0:
plot.set_axis_title(BasePlot.Y_LEFT, 'Correlation')
if per_second:
plot.set_axis_unit(BasePlot.Y_LEFT, 'count/second')
else:
plot.set_axis_unit(BasePlot.Y_LEFT, 'count/segment')
win.add_plot_widget(pW, i, column=i % columns)
win.add_custom_curve_tools()
progress.done()
win.add_legend_option(legends, True)
win.show()
if len(crlgs) > 1:
win.add_x_synchronization_option(True, range(len(crlgs)))
win.add_y_synchronization_option(False, range(len(crlgs)))
return win
def signals(signals,
events=None,
epochs=None,
spike_trains=None,
spikes=None,
show_waveforms=True,
use_subplots=True,
subplot_names=True,
time_unit=pq.s,
y_unit=None,
progress=None):
""" Create a plot from a list of analog signals.
:param list signals: The list of :class:`neo.core.AnalogSignal` objects
to plot.
:param sequence events: A list of Event objects to be included in the
plot.
:param sequence epochs: A list of Epoch objects to be included in the
plot.
:param list spike_trains: A list of :class:`neo.core.SpikeTrain` objects
to be included in the plot. The ``unit`` property (if it exists) is
used for color and legend entries.
:param list spikes: A list :class:`neo.core.Spike` objects to be included
in the plot. The ``unit`` property (if it exists) is used for color
and legend entries.
:param bool show_waveforms: Determines if spikes from
:class:`neo.core.Spike` and :class:`neo.core.SpikeTrain` objects are
shown as waveforms (if available) or vertical lines.
:param bool use_subplots: Determines if a separate subplot for is created
each signal.
:param bool subplot_names: Only valid if ``use_subplots`` is True.
Determines if signal (or channel) names are shown for subplots.
:param Quantity time_unit: The unit of the x axis.
:param progress: Set this parameter to report progress.
:type progress: :class:`spykeutils.progress_indicator.ProgressIndicator`
"""
if not signals:
raise SpykeException(
'Cannot create signal plot: No signal data provided!')
if not progress:
progress = ProgressIndicator()
# Plot title
win_title = 'Analog Signal'
if len(set((s.recordingchannel for s in signals))) == 1:
if signals[0].recordingchannel and signals[0].recordingchannel.name:
win_title += ' | Recording Channel: %s' %\
signals[0].recordingchannel.name
if len(set((s.segment for s in signals))) == 1:
if signals[0].segment and signals[0].segment.name:
win_title += ' | Segment: %s' % signals[0].segment.name
win = PlotDialog(toolbar=True, wintitle=win_title)
if events is None:
events = []
if epochs is None:
epochs = []
if spike_trains is None:
spike_trains = []
if spikes is None:
spikes = []
if show_waveforms:
for st in spike_trains:
if st.waveforms is not None:
spikes.extend(conversions.spike_train_to_spikes(st))
spike_trains = []
else:
unit_spikes = {}
for s in spikes:
unit_spikes.setdefault(s.unit, []).append(s)
for sps in unit_spikes.itervalues():
spike_trains.append(conversions.spikes_to_spike_train(sps, False))
spikes = []
channels = range(len(signals))
channel_indices = []
for s in signals:
if not s.recordingchannel:
channel_indices.append(-1)
else:
channel_indices.append(s.recordingchannel.index)
# Heuristic: If multiple channels have the same index, use channel order
# as index for spike waveforms
nonindices = max(0, channel_indices.count(-1) - 1)
if len(set(channel_indices)) != len(channel_indices) - nonindices:
channel_indices = range(len(signals))
progress.set_ticks((len(spike_trains) + len(spikes) + 1) * len(channels))
offset = 0 * signals[0].units
if use_subplots:
plot = None
for c in channels:
pW = BaseCurveWidget(win)
plot = pW.plot
if subplot_names:
if signals[c].name:
win.set_plot_title(plot, signals[c].name)
elif signals[c].recordingchannel:
if signals[c].recordingchannel.name:
win.set_plot_title(plot,
signals[c].recordingchannel.name)
sample = (1 / signals[c].sampling_rate).simplified
x = (sp.arange(signals[c].shape[0])) * sample + signals[c].t_start
x.units = time_unit
helper.add_epochs(plot, epochs, x.units)
if y_unit is not None:
plot.add_item(make.curve(x, signals[c].rescale(y_unit)))
else:
plot.add_item(make.curve(x, signals[c]))
helper.add_events(plot, events, x.units)
_add_spike_waveforms(plot, spikes, x.units, channel_indices[c],
offset, progress)
for train in spike_trains:
color = helper.get_object_color(train.unit)
helper.add_spikes(plot, train, color, units=x.units)
progress.step()
win.add_plot_widget(pW, c)
plot.set_axis_unit(BasePlot.Y_LEFT,
signals[c].dimensionality.string)
progress.step()
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM, x.dimensionality.string)
else:
channels.reverse()
pW = BaseCurveWidget(win)
plot = pW.plot
helper.add_epochs(plot, epochs, time_unit)
# Find plot y offset
max_offset = 0 * signals[0].units
for i, c in enumerate(channels[1:], 1):
cur_offset = signals[channels[i - 1]].max() - signals[c].min()
if cur_offset > max_offset:
max_offset = cur_offset
offset -= signals[channels[0]].min()
for c in channels:
sample = (1 / signals[c].sampling_rate).simplified
x = (sp.arange(signals[c].shape[0])) * sample + signals[c].t_start
x.units = time_unit
if y_unit is not None:
plot.add_item(
make.curve(x, (signals[c] + offset).rescale(y_unit)))
else:
plot.add_item(make.curve(x, signals[c] + offset))
_add_spike_waveforms(plot, spikes, x.units, channel_indices[c],
offset, progress)
offset += max_offset
progress.step()
helper.add_events(plot, events, x.units)
for train in spike_trains:
color = helper.get_object_color(train.unit)
helper.add_spikes(plot, train, color, units=x.units)
progress.step()
win.add_plot_widget(pW, 0)
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM, x.dimensionality.string)
plot.set_axis_unit(BasePlot.Y_LEFT, signals[0].dimensionality.string)
win.add_custom_curve_tools()
units = set([s.unit for s in spike_trains])
units = units.union([s.unit for s in spikes])
progress.done()
helper.make_window_legend(win, units, False)
win.show()
if use_subplots:
win.add_x_synchronization_option(True, channels)
win.add_y_synchronization_option(False, channels)
return win
def isi(trains, bin_size, cut_off, bar_plot=False, time_unit=pq.ms):
""" Create a plot dialog with an interspike interval histogram.
:param dict trains: Dictionary with lists of spike trains indexed by
units for which to display ISI histograms
:param bin_size: Bin size (time)
:type bin_size: Quantity scalar
:param cut_off: End of histogram (time)
:type bin_size: Quantity scalar
:param bool bar_plot: If ``True``, create a bar ISI histogram for each
index in ``trains``. Else, create a line ISI histogram.
:param Quantity time_unit: Unit of X-Axis.
"""
if not trains:
raise SpykeException('No spike trains for ISI histogram')
win_title = 'ISI Histogram | Bin size: ' + str(bin_size)
win = PlotDialog(toolbar=True,
wintitle=win_title,
min_plot_width=150,
min_plot_height=100)
bin_size = bin_size.rescale(time_unit)
cut_off = cut_off.rescale(time_unit)
bins = sp.arange(0 * time_unit, cut_off, bin_size) * time_unit
legends = []
if bar_plot:
ind = 0
columns = int(sp.sqrt(len(trains)))
for i, train_list in trains.iteritems():
pW = BaseCurveWidget(win)
plot = pW.plot
intervals = []
for t in train_list:
t = t.rescale(time_unit)
sTrain = sp.asarray(t)
sTrain.sort()
intervals.extend(sp.diff(sTrain))
(isi, bins) = sp.histogram(intervals, bins)
if i and hasattr(i, 'name') and i.name:
name = i.name
else:
name = 'Unknown'
show_isi = list(isi)
show_isi.insert(0, show_isi[0])
curve = make.curve(bins,
show_isi,
name,
color='k',
curvestyle="Steps",
shade=1.0)
plot.add_item(curve)
# Create legend
color = helper.get_object_color(i)
color_curve = make.curve([], [],
name,
color,
'NoPen',
linewidth=1,
marker='Rect',
markerfacecolor=color,
markeredgecolor=color)
plot.add_item(color_curve)
legends.append(make.legend(restrict_items=[color_curve]))
plot.add_item(legends[-1])
# Prepare plot
plot.set_antialiasing(False)
scale = plot.axisScaleDiv(BasePlot.Y_LEFT)
plot.setAxisScale(BasePlot.Y_LEFT, 0, scale.upperBound())
if ind % columns == 0:
plot.set_axis_title(BasePlot.Y_LEFT, 'Number of intervals')
if ind >= len(trains) - columns:
plot.set_axis_title(BasePlot.X_BOTTOM, 'Interval length')
plot.set_axis_unit(BasePlot.X_BOTTOM,
time_unit.dimensionality.string)
win.add_plot_widget(pW, ind, column=ind % columns)
ind += 1
else:
pW = BaseCurveWidget(win)
plot = pW.plot
legend_items = []
for i, train_list in trains.iteritems():
intervals = []
for t in train_list:
t = t.rescale(time_unit)
sTrain = sp.asarray(t)
sTrain.sort()
intervals.extend(sp.diff(sTrain))
(isi, bins) = sp.histogram(intervals, bins)
if i and hasattr(i, 'name') and i.name:
name = i.name
else:
name = 'Unknown'
color = helper.get_object_color(i)
curve = make.curve(bins, isi, name, color=color)
legend_items.append(curve)
plot.add_item(curve)
win.add_plot_widget(pW, 0)
legends.append(make.legend(restrict_items=legend_items))
plot.add_item(legends[-1])
plot.set_antialiasing(True)
plot.set_axis_title(BasePlot.Y_LEFT, 'Number of intervals')
plot.set_axis_title(BasePlot.X_BOTTOM, 'Interval length')
plot.set_axis_unit(BasePlot.X_BOTTOM, time_unit.dimensionality.string)
win.add_custom_curve_tools()
win.add_legend_option(legends, True)
win.show()
if bar_plot and len(trains) > 1:
win.add_x_synchronization_option(True, range(len(trains)))
win.add_y_synchronization_option(False, range(len(trains)))
return win
def psth(trains, events=None, start=0 * pq.ms, stop=None,
bin_size=100 * pq.ms, rate_correction=True, bar_plot=False,
time_unit=pq.ms, progress=None):
""" Create a peri stimulus time histogram.
The peri stimulus time histogram gives an estimate of the instantaneous
rate.
:param dict trains: A dictionary of :class:`neo.core.SpikeTrain` lists.
:param dict events: A dictionary of Event objects, indexed by segment.
The events will be at time 0 on the plot. If None, spike trains
are used unmodified.
:param start: The desired time for the start of the first bin. It
will be recalculated if there are spike trains which start later
than this time. This parameter can be negative (which could be
useful when aligning on events).
:type start: Quantity scalar
:param stop: The desired time for the end of the last bin. It will
be recalculated if there are spike trains which end earlier
than this time.
:type stop: Quantity scalar
:param bin_size: The bin size for the histogram.
:type bin_size: Quantity scalar
:param bool rate_correction: Determines if a rates (``True``) or
counts (``False``) are shown.
:param bool bar_plot: Determines if a bar plot (``True``) or a line
plot (``False``) will be created. In case of a bar plot, one plot
for each index in ``trains`` will be created.
:param Quantity time_unit: Unit of X-Axis.
:param progress: Set this parameter to report progress.
:type progress: :class:`spykeutils.progress_indicator.ProgressIndicator`
"""
if not trains:
raise SpykeException('No spike trains for PSTH!')
if not progress:
progress = ProgressIndicator()
# Align spike trains
for u in trains:
if events:
trains[u] = rate_estimation.aligned_spike_trains(
trains[u], events)
rates, bins = rate_estimation.psth(
trains, bin_size, start=start, stop=stop,
rate_correction=rate_correction)
bins = bins.rescale(time_unit)
if not psth:
raise SpykeException('No spike trains for PSTH!')
win_title = 'PSTH | Bin size %.2f %s' % (bin_size,
time_unit.dimensionality.string)
win = PlotDialog(toolbar=True, wintitle=win_title, min_plot_width=150,
min_plot_height=100)
legends = []
if bar_plot:
ind = 0
columns = int(sp.sqrt(len(rates)))
for i, r in rates.iteritems():
if i and hasattr(i, 'name') and i.name:
name = i.name
else:
name = 'Unknown'
pW = BaseCurveWidget(win)
plot = pW.plot
show_rates = list(r)
show_rates.insert(0, show_rates[0])
curve = make.curve(
bins, show_rates, name, color='k',
curvestyle="Steps", shade=1.0)
plot.add_item(curve)
# Create legend
color = helper.get_object_color(i)
color_curve = make.curve(
[], [], name, color, 'NoPen', linewidth=1, marker='Rect',
markerfacecolor=color, markeredgecolor=color)
plot.add_item(color_curve)
legends.append(make.legend(restrict_items=[color_curve]))
plot.add_item(legends[-1])
# Prepare plot
plot.set_antialiasing(False)
scale = plot.axisScaleDiv(BasePlot.Y_LEFT)
plot.setAxisScale(BasePlot.Y_LEFT, 0, scale.upperBound())
if ind % columns == 0:
if not rate_correction:
plot.set_axis_title(BasePlot.Y_LEFT, 'Spike Count')
else:
plot.set_axis_title(BasePlot.Y_LEFT, 'Rate')
plot.set_axis_unit(BasePlot.Y_LEFT, 'Hz')
if ind >= len(trains) - columns:
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(
BasePlot.X_BOTTOM, time_unit.dimensionality.string)
win.add_plot_widget(pW, ind, column=ind % columns)
ind += 1
else:
bins = 0.5 * sp.diff(bins) + bins[:-1]
pW = BaseCurveWidget(win)
plot = pW.plot
legend_items = []
for i, r in rates.iteritems():
if i and hasattr(i, 'name') and i.name:
name = i.name
else:
name = 'Unknown'
curve = make.curve(
bins, r, name,
color=helper.get_object_color(i))
legend_items.append(curve)
plot.add_item(curve)
win.add_plot_widget(pW, 0)
legends.append(make.legend(restrict_items=legend_items))
plot.add_item(legends[-1])
if not rate_correction:
plot.set_axis_title(BasePlot.Y_LEFT, 'Spike Count')
else:
plot.set_axis_title(BasePlot.Y_LEFT, 'Rate')
plot.set_axis_unit(BasePlot.Y_LEFT, 'Hz')
plot.set_axis_title(BasePlot.X_BOTTOM, 'Time')
plot.set_axis_unit(BasePlot.X_BOTTOM, time_unit.dimensionality.string)
plot.set_antialiasing(True)
win.add_custom_curve_tools()
win.add_legend_option(legends, True)
progress.done()
win.show()
if bar_plot and len(rates) > 1:
win.add_x_synchronization_option(True, range(len(rates)))
win.add_y_synchronization_option(False, range(len(rates)))
return win