class StackViewer(object):
"""
Parameters
----------
viewer : object
expected to have update_image method and fig attribute
images : array-like
must support integer indexing and return a 2D array
"""
def __init__(self, viewer, images):
self.viewer = viewer
self.images = images
length = len(self.images)
fig = self.viewer._fig
slider_ax = fig.add_axes([0.1, 0.01, 0.8, 0.02])
self.slider = Slider(slider_ax,
'Frame',
0,
length - 1,
0,
valfmt='%d/{}'.format(length - 1))
self.slider.on_changed(self.update)
self.update(0) # Trigger the initialization of viewer.
def update(self, val):
if not isinstance(val, int):
self.slider.set_val(int(round(val)))
# sends up through 'update' again
self.viewer.update_image(self.images[int(val)])
def generate_plots():
global rts_filename, img3d, contour_matrix, im, structure_wanted, list_of_ct_filenames, current_slice_slider,contour_plot
# get all the relevant data (choose random structure set, get ct and contour data from dicoms)
structure_wanted, rts_filename, list_of_ct_filenames = get_random_structure_set()
img3d, contour_matrix, x_origin, y_origin, pixel_size = prepare_dicom_data_for_individual_case(rts_filename, list_of_ct_filenames, structure_wanted)
#smooth the human contours if desired
if smooth_human_contours == True and human_rts_filename in rts_filename:
smooth_contour_matrix()
# can remove ax.clear() for randomly generated colors of contours
ax.clear()
# deal with the Slider
# try statement only runs if it's NOT the first image plotted (ie slider doesn't exist yet, needs to be initialized)
# subsequent images just update the slider rather than re-creating it
try:
current_slice_slider.set_val(int(img3d.shape[0]/2))
current_slice_slider.valmax = img3d.shape[0]-1
current_slice_slider.ax.set_xlim(current_slice_slider.valmin, current_slice_slider.valmax)
except NameError:
current_slice_slider = Slider(slice_slider_axes, 'CT Slice', 0, img3d.shape[0]-1, valstep=1, valfmt='%0.0f')
current_slice_slider.set_val(int(img3d.shape[0]/2))
current_slice_slider.on_changed(on_slider_change)
#plot the data, and connect
im = ax.imshow(img3d[int(current_slice_slider.val), :, :],
extent=[x_origin, x_origin + img3d.shape[1] * pixel_size, y_origin, y_origin + img3d.shape[2] * pixel_size],
cmap='Greys_r', vmin=hu_min, vmax=hu_max, animated = True, interpolation = 'nearest', origin = 'upper')
contour_plot = ax.plot(contour_matrix[int(current_slice_slider.val)][0], contour_matrix[int(current_slice_slider.val)][1] )
def main(filenames, equal_axes):
all_data = []
tmin = tmax = 0
xmin, ymin, xmax, ymax = 1e100, 1e100, -1e100, -1e100
for i, filename in enumerate(filenames):
name = str(i + 1)
if ':' in filename:
filename, name = filename.split(':')
print('Reading %s from file %s' % (name, filename))
_description, _value, _dim, timesteps, data = read_iso_surfaces(
filename)
all_data.append((name, numpy.array(timesteps), data))
tmax = max(tmax, timesteps[-1])
for contours in data:
for contour in contours:
xmin = min(xmin, numpy.min(contour[0]))
ymin = min(ymin, numpy.min(contour[1]))
xmax = max(xmax, numpy.max(contour[0]))
ymax = max(ymax, numpy.max(contour[1]))
fig, ax = pyplot.subplots()
pyplot.subplots_adjust(bottom=0.25)
axcolor = '#a1b8dd'
slider_ax = pyplot.axes([0.1, 0.1, 0.8, 0.03], facecolor=axcolor)
slider = Slider(slider_ax, 'Time', tmin, tmax, valinit=tmin)
xdiff = xmax - xmin
ydiff = ymax - ymin
xmin, xmax = xmin - 0.1 * xdiff, xmax + 0.1 * xdiff
ymin, ymax = ymin - 0.1 * ydiff, ymax + 0.1 * ydiff
ax.set_xlim(xmin, xmax)
ax.set_ylim(ymin, ymax)
def update(val):
xmin, xmax = ax.get_xlim()
ymin, ymax = ax.get_ylim()
ax.clear()
t = slider.val
for name, timesteps, data in all_data:
i = numpy.argmin(abs(timesteps - t))
dt = timesteps[1] - timesteps[0]
if abs(timesteps[i] - t) > 1.5 * dt:
continue
plotit(ax, data[i], name)
ax.set_xlim(xmin, xmax)
ax.set_ylim(ymin, ymax)
if equal_axes:
ax.set_aspect('equal')
if len(filenames) > 1:
ax.legend(loc='lower right')
fig.canvas.draw_idle()
slider.on_changed(update)
slider.set_val(tmin)
pyplot.show()
class ImgView(object):
def __init__(self, img, fig_id=None, imin=None, imax=None):
self.fig = plt.figure(fig_id)
self.ax = self.fig.add_axes([0.1, 0.25, 0.8, 0.7])
self.fig.subplots_adjust(bottom=0.25)
axcolor = 'lightgoldenrodyellow'
self.axlo = self.fig.add_axes([0.15, 0.1, 0.65, 0.03], axisbg=axcolor)
self.axhi = self.fig.add_axes([0.15, 0.15, 0.65, 0.03], axisbg=axcolor)
self.axrefr = self.fig.add_axes([0.15, 0.05, 0.10, 0.03],
axisbg=axcolor)
imsort = np.sort(img.flatten())
n = len(imsort)
if imin is None:
imin = imsort[int(n * 0.005)]
if imax is None:
imax = imsort[int(n * 0.995)]
self.slo = Slider(self.axlo,
'Scale min',
imin,
imax,
valinit=imsort[int(n * 0.02)])
self.shi = Slider(self.axhi,
'Scale max',
imin,
imax,
valinit=imsort[int(n * 0.98)])
self.brefr = Button(self.axrefr, 'Refresh')
self.slo.on_changed(self.update_slider)
self.shi.on_changed(self.update_slider)
self.brefr.on_clicked(self.refresh)
self.set_img(img)
def update_slider(self, val):
if self.shi.val <= self.slo.val:
self.shi.set_val(self.slo.val + 1)
self.imgplt.set_clim(self.slo.val, self.shi.val)
self.fig.canvas.draw()
def set_img(self, img=None):
if img is not None:
self.img = img
self.ax.set_xlim(0, self.img.shape[1] - 1)
self.ax.set_ylim(0, self.img.shape[0] - 1)
self.imgplt = self.ax.imshow(self.img,
vmin=self.slo.val,
vmax=self.shi.val,
cmap='hot',
origin='lower',
interpolation='nearest')
self.imgplt.set_cmap('hot')
divider = make_axes_locatable(self.ax)
cax = divider.append_axes("right", size="5%", pad=0.05)
self.fig.colorbar(self.imgplt, cax=cax)
self.fig.canvas.draw()
def refresh(self, event=None):
self.set_img()
class GUI:
def __init__(self, coinToss):
self._coinToss = coinToss
maxTosses = coinToss.tosses
axcolor = 'lightgoldenrodyellow'
self.figure = pyplot.figure()
self.mainAxis = pyplot.axes([0.05, 0.2, 0.9, 0.75])
# Slider for number of tosses
tossAxis = pyplot.axes([0.1, 0.05, 0.8, 0.05])
self.tossSlider = Slider(tossAxis,
'Tosses',
0.,
1. * maxTosses,
valinit=0,
valfmt=u'%d')
self.tossSlider.on_changed(lambda x: self.draw(x))
# Reset button
resetAxis = pyplot.axes([0.8, 0.85, 0.1, 0.05])
self.resetButton = Button(resetAxis,
'Re-toss',
color=axcolor,
hovercolor='0.975')
self.resetButton.on_clicked(self.retoss)
# Key press events
self.figure.canvas.mpl_connect('key_press_event',
lambda x: self.press(x))
self._coinToss.doTosses()
def retoss(self, event):
self._coinToss.doTosses()
self.draw(self.tossSlider.val)
def press(self, event):
if event.key == u'left' and self.tossSlider.val > self.tossSlider.valmin:
self.tossSlider.set_val(self.tossSlider.val - 1)
if event.key == u'right' and self.tossSlider.val < self.tossSlider.valmax:
self.tossSlider.set_val(self.tossSlider.val + 1)
if event.key == u'r':
self.retoss(event)
def draw(self, x=0):
c = self._coinToss
m = max(enumerate(c.posterior[int(x), :]), key=operator.itemgetter(1))
self.mainAxis.clear()
self.mainAxis.plot(c.conditional,
c.posterior[int(x), :],
lw=2,
color='red')
self.mainAxis.vlines(c.conditional[m[0]], 0, c.posterior.max())
self.figure.canvas.draw()
class Index(object):
def __init__(self, ax_slider, ax_prev, ax_next):
self.ind = 0
self.num = len(wavelengths)
self.bnext = Button(ax_next, 'Next')
self.bnext.on_clicked(self.next)
self.bprev = Button(ax_prev, 'Previous')
self.bprev.on_clicked(self.prev)
self.slider = Slider(ax_slider,
"Energy Resolution: {:.2f} nm".format(
wavelengths[0]),
0,
self.num,
valinit=0,
valfmt='%d')
self.slider.valtext.set_visible(False)
self.slider.label.set_horizontalalignment('center')
self.slider.on_changed(self.update)
position = ax_slider.get_position()
self.slider.label.set_position((0.5, -0.5))
self.slider.valtext.set_position((0.5, -0.5))
def next(self, event):
i = (self.ind + 1) % (self.num + 1)
self.slider.set_val(i)
def prev(self, event):
i = (self.ind - 1) % (self.num + 1)
self.slider.set_val(i)
def update(self, i):
self.ind = int(i)
image.set_data(R[self.ind])
if self.ind != len(wavelengths):
self.slider.label.set_text(
"Energy Resolution: {:.2f} nm".format(
wavelengths[self.ind]))
else:
self.slider.label.set_text("Calibrated Pixels")
if self.ind != len(wavelengths):
number = 11
cbar.set_clim(vmin=0, vmax=maximum)
cbar_ticks = np.linspace(0.,
maximum,
num=number,
endpoint=True)
else:
number = 2
cbar.set_clim(vmin=0, vmax=1)
cbar_ticks = np.linspace(0., 1, num=number)
cbar.set_ticks(cbar_ticks)
cbar.draw_all()
plt.draw()
def adv_window():
figs, axx = plt.subplots(num = 'Advanced settings')
figs.canvas.mpl_connect('key_press_event', adv_exit)
axx.axis('off')
#bx1_as = plt.axes([0.05, 0.3, 0.15, 0.11])
#bx1_as.set_axis_off()
#button_as1 = CheckButtons(bx1_as, ['lambda1'], [1])
axx.axis('off')
ax_as1 = plt.axes([0.15, 0.02, 0.5, 0.05])
slider_as1 = Slider(ax_as1, 'lambda1', 0.1, 4, dragging = True, valstep = 0.1)
ax_as2 = plt.axes([0.15, 0.10, 0.5, 0.05])
slider_as2 = Slider(ax_as2, 'lambda2', 0.1, 4, dragging = True, valstep = 0.1)
ax_as3 = plt.axes([0.15, 0.18, 0.5, 0.05])
slider_as3 = Slider(ax_as3, 'smoothing', 0, 4, dragging = True, valstep = 1)
ax_as4 = plt.axes([0.15, 0.26, 0.5, 0.05])
slider_as4 = Slider(ax_as4, 'iterations', 1, 1000, dragging = True, valstep = 1)
ax_as5 = plt.axes([0.15, 0.34, 0.5, 0.05])
slider_as5 = Slider(ax_as5, 'radius', 0.5, 5, dragging = True, valstep = 0.1)
ax_b1 = plt.axes([0.85, 0.15, 0.07, 0.08])
ax_b2 = plt.axes([0.85, 0.05, 0.07, 0.08])
but_as1 = Button(ax_b1, 'exit', color = 'beige', hovercolor = 'beige')
but_as2 = Button(ax_b2, 'start', color = 'beige', hovercolor = 'beige')
#ax_textbox = plt.axes([0, 0.4, 0.5, 0.4])
#axx.axis('off')
textstr = "Press ENTER in terminal to start segmentation. \n Shouldn't be necessairy to change settings below, but can be tuned if \n resulting segmentation is not ideal. \n Especially if small nodule: try setting lambda1 <= lambda2 \n Or if very nonhomogeneous nodule: try setting lambda1 > lambda2."
props = dict(boxstyle='round', facecolor='wheat')
axx.text(-0.18, 0.25, textstr, transform=ax.transAxes, fontsize=12,
verticalalignment='top', bbox=props)
slider_as1.set_val(lambda1)
slider_as2.set_val(lambda2)
slider_as3.set_val(smoothing)
slider_as4.set_val(iterations)
slider_as5.set_val(rad)
but_as1.on_clicked(adv_exit)
but_as2.on_clicked(adv_start)
figs.canvas.draw_idle()
return figs, axx, slider_as1, slider_as2, slider_as3, slider_as4, slider_as5
def plot_hfo_samples(hfo_detection_run: HfoDetectionRun):
periods = hfo_detection_run.detector.last_run.analytics.periods
fig_height = 6
fig_width = 10
rows = 4
columns = 1
fig = plt.figure(figsize=(fig_width, fig_height))
plt.rc('font', family='sans-serif')
spec = gridspec.GridSpec(rows, columns, figure=fig, hspace=0.7)
bandwidth_axes = fig.add_subplot(spec[0, 0])
spike_train_axes = fig.add_subplot(spec[1, 0])
raster_axes = fig.add_subplot(spec[2, 0])
slider_axes = fig.add_subplot(8, 1, 8)
period_windows = list(zip(periods.start, periods.stop))
if len(period_windows) == 0:
return
slider = Slider(slider_axes,
'Period Index\n(Interactive)',
1,
len(period_windows) + 1,
valinit=1,
valstep=1.0)
initial_start, initial_stop = period_windows[0]
_plot_hfo_sample(hfo_detection_run, np.float64(initial_start),
np.float64(initial_stop), bandwidth_axes,
spike_train_axes, raster_axes)
def plot_time(one_based_index):
start, stop = period_windows[int(np.round(one_based_index - 1))]
_plot_hfo_sample(hfo_detection_run, np.float64(start),
np.float64(stop), bandwidth_axes, spike_train_axes,
raster_axes)
fig.canvas.draw_idle()
slider.on_changed(plot_time)
if should_show_plot(hfo_detection_run.configuration):
plt.show()
if should_save_plot(hfo_detection_run.configuration):
for one_based_index in range(1, len(period_windows) + 1):
slider.set_val(one_based_index)
save_or_show_channel_plot(f'hfo_sample_period_{one_based_index}',
hfo_detection_run)
class HistoryPlotter:
def __init__(self,saved_filename):
self.plotter = RealTimePlotter()
self.plotter.fig.suptitle('History data', fontsize='14', fontweight='bold')
self.plotter.fig.subplots_adjust(bottom=0.23,hspace=0.5)
self.index = 0
self.saved_filename = saved_filename
self.data = []
self.prev_axis = plt.axes([0.395, 0.03, 0.1, 0.06])
self.next_axis = plt.axes([0.505, 0.03, 0.1, 0.06])
self.btn_next = Button(self.next_axis, 'Next')
self.btn_next.on_clicked(self.next)
self.btn_prev = Button(self.prev_axis, 'Previous')
self.btn_prev.on_clicked(self.prev)
self.slider_axis = plt.axes([0.25, 0.11, 0.5, 0.03])
self.read_data()
self.slider = Slider(self.slider_axis, 'chunk',0,len(self.data)-1, valinit=0,valfmt='%10.0f')
self.slider.on_changed(self.update)
def read_data(self):
with bz2.BZ2File(self.saved_filename,'r') as f:
self.data = pickle.load(f)
print('Number of chunks',len(self.data) )
def loop(self):
try:
self.plotter.show()
self.plotter.plot(self.data[self.index]['buffer'],self.data[self.index]['stats'])
plt.show(block=True)
except KeyboardInterrupt:
print("Stopping...")
def update(self,val):
self.index=int(val)
self.plotter.plot(self.data[self.index]['buffer'],self.data[self.index]['stats'])
def next(self,event):
if(self.index < len(self.data)-1):
self.index +=1
self.plotter.plot(self.data[self.index]['buffer'],self.data[self.index]['stats'])
self.slider.set_val(self.index)
def prev(self,event):
if(self.index >0):
self.index -=1
self.plotter.plot(self.data[self.index]['buffer'],self.data[self.index]['stats'])
self.slider.set_val(self.index)
class ImgView(object):
def __init__(self, img, fig_id=None, imin=None, imax=None):
self.fig = plt.figure(fig_id)
self.ax = self.fig.add_axes([0.1, 0.25, 0.8, 0.7])
self.fig.subplots_adjust(bottom=0.25)
axcolor = 'lightgoldenrodyellow'
self.axlo = self.fig.add_axes([0.15, 0.1, 0.65, 0.03], axisbg=axcolor)
self.axhi = self.fig.add_axes([0.15, 0.15, 0.65, 0.03], axisbg=axcolor)
self.axrefr = self.fig.add_axes([0.15, 0.05, 0.10, 0.03], axisbg=axcolor)
imsort = np.sort(img.flatten())
n = len(imsort)
if imin is None:
imin = imsort[int(n * 0.005)]
if imax is None:
imax = imsort[int(n * 0.995)]
self.slo = Slider(self.axlo, 'Scale min', imin, imax, valinit=imsort[int(n * 0.02)])
self.shi = Slider(self.axhi, 'Scale max', imin, imax, valinit=imsort[int(n * 0.98)])
self.brefr = Button(self.axrefr, 'Refresh')
self.slo.on_changed(self.update_slider)
self.shi.on_changed(self.update_slider)
self.brefr.on_clicked(self.refresh)
self.set_img(img)
def update_slider(self, val):
if self.shi.val <= self.slo.val:
self.shi.set_val(self.slo.val + 1)
self.imgplt.set_clim(self.slo.val, self.shi.val)
self.fig.canvas.draw()
def set_img(self, img=None):
if img is not None:
self.img = img
self.ax.set_xlim(0, self.img.shape[1]-1)
self.ax.set_ylim(0, self.img.shape[0]-1)
self.imgplt = self.ax.imshow(self.img, vmin=self.slo.val, vmax=self.shi.val,
cmap='hot', origin='lower',
interpolation='nearest')
self.imgplt.set_cmap('hot')
divider = make_axes_locatable(self.ax)
cax = divider.append_axes("right", size="5%", pad=0.05)
self.fig.colorbar(self.imgplt, cax=cax)
self.fig.canvas.draw()
def refresh(self, event=None):
self.set_img()
class Index(object):
def __init__(self, ax_slider, ax_prev, ax_next):
self.ind = 0
self.num = len(wavelengths)
self.bnext = Button(ax_next, 'Next')
self.bnext.on_clicked(self.next)
self.bprev = Button(ax_prev, 'Previous')
self.bprev.on_clicked(self.prev)
self.slider = Slider(ax_slider,
"Energy Resolution: {:.2f} nm".format(wavelengths[0]), 0,
self.num, valinit=0, valfmt='%d')
self.slider.valtext.set_visible(False)
self.slider.label.set_horizontalalignment('center')
self.slider.on_changed(self.update)
position = ax_slider.get_position()
self.slider.label.set_position((0.5, -0.5))
self.slider.valtext.set_position((0.5, -0.5))
def next(self, event):
i = (self.ind + 1) % (self.num + 1)
self.slider.set_val(i)
def prev(self, event):
i = (self.ind - 1) % (self.num + 1)
self.slider.set_val(i)
def update(self, i):
self.ind = int(i)
image.set_data(R[self.ind])
if self.ind != len(wavelengths):
self.slider.label.set_text("Energy Resolution: {:.2f} nm"
.format(wavelengths[self.ind]))
else:
self.slider.label.set_text("Calibrated Pixels")
if self.ind != len(wavelengths):
number = 11
cbar.set_clim(vmin=0, vmax=maximum)
cbar_ticks = np.linspace(0., maximum, num=number, endpoint=True)
else:
number = 2
cbar.set_clim(vmin=0, vmax=1)
cbar_ticks = np.linspace(0., 1, num=number)
cbar.set_ticks(cbar_ticks)
cbar.draw_all()
plt.draw()
def set_val(self, val):
"""
Set the value and update the color.
Notes
-----
valmin/valmax are set on the parent to 0 and len(depths).
"""
val = int(val)
# valmax is not allowed, since it is out of the array.
# valmin is allowed since 0 index is in depth array.
if val < self.valmin or val >= self.valmax:
# invalid, so ignore
return
# activate color is first since we still have access to self.val
self.updatePageDepthColor(val)
Slider.set_val(self, val)
class GUI:
def __init__(self, coinToss):
self._coinToss = coinToss
maxTosses = coinToss.tosses
axcolor = 'lightgoldenrodyellow'
self.figure = pyplot.figure()
self.mainAxis = pyplot.axes([0.05, 0.2, 0.9, 0.75])
# Slider for number of tosses
tossAxis = pyplot.axes([0.1, 0.05, 0.8, 0.05])
self.tossSlider = Slider(tossAxis, 'Tosses', 0., 1.*maxTosses, valinit=0, valfmt=u'%d')
self.tossSlider.on_changed(lambda x: self.draw(x))
# Reset button
resetAxis = pyplot.axes([0.8, 0.85, 0.1, 0.05])
self.resetButton = Button(resetAxis, 'Re-toss', color=axcolor, hovercolor='0.975')
self.resetButton.on_clicked(self.retoss)
# Key press events
self.figure.canvas.mpl_connect('key_press_event', lambda x: self.press(x))
self._coinToss.doTosses()
def retoss(self, event):
self._coinToss.doTosses()
self.draw(self.tossSlider.val)
def press(self, event):
if event.key == u'left' and self.tossSlider.val > self.tossSlider.valmin:
self.tossSlider.set_val(self.tossSlider.val - 1)
if event.key == u'right' and self.tossSlider.val < self.tossSlider.valmax:
self.tossSlider.set_val(self.tossSlider.val + 1)
if event.key == u'r':
self.retoss(event)
def draw(self, x = 0):
c = self._coinToss
m = max(enumerate(c.posterior[int(x),:]), key=operator.itemgetter(1))
self.mainAxis.clear()
self.mainAxis.plot(c.conditional,c.posterior[int(x),:], lw=2, color='red')
self.mainAxis.vlines(c.conditional[m[0]],0,c.posterior.max())
self.figure.canvas.draw()
class IndexTracker(object):
def __init__(self, ax, X, step=41):
self.ax = ax
ax.figure.subplots_adjust(left=0.25, bottom=0.25)
ax.set_title('use scroll wheel to navigate images')
self.step = step
self.X = X
self.slices, rows, cols = X.shape
self.ind = 0
self.im = ax.imshow(self.X[self.ind, :, :],
vmin=np.min(X),
vmax=np.max(X))
ax = fig.add_axes([0.25, 0.1, 0.65, 0.03])
self.slider = Slider(ax,
'Axis %i index' % self.slices,
0,
self.slices,
valinit=self.ind,
valfmt='%i')
self.slider.on_changed(self.update_slider)
self.update()
def press(self, event):
if event.key == 'right':
self.ind = (self.ind + self.step) % self.slices
elif event.key == 'left':
self.ind = (self.ind - self.step) % self.slices
self.slider.set_val(self.ind)
self.update()
def update_slider(self, event):
ind = int(self.slider.val)
self.ind = ind
self.update()
def update(self):
self.im.set_data(self.X[self.ind, :, :].T)
ax.set_ylabel('slice %s' % self.ind)
self.im.axes.figure.canvas.draw()
def plot_data(self):
import matplotlib.pyplot as plt
from matplotlib.widgets import Slider
fig = plt.figure()
ax0 = plt.axes([0.085, 0.2, 0.9, 0.75])
ax1 = plt.axes([0.21, 0.02, 0.7, 0.03])
self.draw_vals(ax0)
B = Slider(ax1,
label='Beta value (1e-8)',
valmin=4,
valmax=50,
valstep=0.1)
Slider.set_val(B, 10)
Slider.on_changed(B, self.update_beta)
plt.show()
def animate_plot_to_pictures(min_speed, max_speed, pictures):
axcolor = 'lightgoldenrodyellow'
axspeed = plt.axes([0.125, 0.05, 0.65, 0.03], facecolor=axcolor)
sspeed = Slider(axspeed, 'Speed', 0, max_speed, valinit=0, valstep=1)
for i in range(pictures):
speed = 1.0 * i / (pictures-1) * max_speed + min_speed
accels, rads = strafevis.strafe_stats.get_stats(720, strafevis.strafe_stats.StatType.ACCEL, speed=speed)
display_axes = plt.subplot(1, 1, 1, polar=True)
norm = mpl.colors.Normalize(0.0, 2 * np.pi)
cmap = AngleMap(accels)
cb = mpl.colorbar.ColorbarBase(display_axes, cmap=cmap,
norm=norm,
orientation='horizontal')
# aesthetics - get rid of border and axis labels
cb.outline.set_visible(False)
display_axes.set_axis_off()
sspeed.set_val(speed)
plt.savefig('pic_%04d.png' % i)
class Visualizer(object):
def __init__(self, snapshot, save_on_close=None):
self.snapshot = snapshot
self.save_on_close = save_on_close
self.T = self.snapshot.human.values()[0].T
self.choice = None
@property
def t(self):
return self._t
@t.setter
def t(self, value):
self._t = value
for scene in self.scenes:
scene.t = self.t
def select(self, key):
self.choice = key
plt.close(self.fig)
def close(self, event):
if self.save_on_close is not None:
self.fig.savefig(self.save_on_close)
def run(self):
self.fig, self.ax = plt.subplots(1, len(self.snapshot.keys()), sharex=True, sharey=True, figsize=(13, 7))
self.fig.canvas.mpl_connect('key_press_event', self.key_press)
self.fig.canvas.mpl_connect('close_event', self.close)
self.scenes = [Scene(ax, self.snapshot.view(key)) for ax, key in zip(self.ax, self.snapshot.keys())]
self.fig.subplots_adjust(bottom=0.15, top=0.85)
box = self.fig.add_axes([0.15, 0.05, 0.7, 0.05])
self.slider = Slider(box, 'Time', 0., self.T, valinit=0.)
self.t = 0.
def update_t(t):
self.t = t
self.slider.on_changed(update_t)
def click(key):
def f(event):
self.select(key)
return f
self.buttons = []
for ax, key in zip(self.ax, self.snapshot.keys()):
box = ax.figbox
box = self.fig.add_axes([box.x0, box.y1+0.05, box.width, 0.05])
self.buttons.append(Button(box, 'Prefer {}'.format(key)))
self.buttons[-1].on_clicked(click(key))
plt.show()
def key_press(self, event):
if event.key=='escape':
plt.close(self.fig)
elif event.key=='r':
self.slider.set_val(0.)
elif event.key=='up':
self.slider.set_val(min(max(self.t+0.2, 0), self.T))
elif event.key=='down':
self.slider.set_val(min(max(self.t-0.2, 0), self.T))
elif event.key.lower() in [s.lower() for s in self.snapshot.keys()]:
for key in self.snapshot.keys():
if event.key.lower()==key.lower():
self.select(key)
class slicer(object):
def __init__(self, data, axis, init_slice):
self.axis = axis
self.axis_label = {0:'X', 1:'Y', 2:'Z'}[axis]
self.data = np.swapaxes(data, self.axis, 0)
self.slice_index = init_slice
self.slice = self.data[self.slice_index]
self.max_index = self.data.shape[0]
self.fig = plt.figure()
self.ax = self.fig.add_subplot(111)
self.ax.imshow(self.slice.T, interpolation='none', origin='lower')
self.along_axis = plt.axes([0.2, 0.1, 0.65, 0.03])
self.slab = Slider(self.along_axis, '%s_Slab'%self.axis_label, 0, self.max_index , valinit=self.slice_index, valfmt='%i')
self.slab.on_changed(self.update_figure)
self.fig.canvas.mpl_connect('key_press_event',self.update_slice_index)
plt.show()
def draw(self):
im = your_function(self.values)
pylab.show()
self.ax.imshow(im)
def update_slice_index(self, event):
if event.key=='+':
self.slice_index += 1
elif event.key == '-':
self.slice_index -= 1
if self.slice_index < 0:
self.slice_index = self.max_index
self.slab.set_val(self.slice_index)
def update_figure(self, event = None):
self.slice_index = int(self.slab.val%self.max_index)
self.slice = self.data[self.slice_index]
self.ax.imshow(self.slice.T, interpolation='none', origin='lower')
self.fig.canvas.draw()
class InteractivePlot(object):
def __init__(self, record_wins, sliding_wins, example_labels,
similarities, is_test=None, plot_rc=None):
self.score_plot = ScorePlot(record_wins, sliding_wins, example_labels,
similarities, is_test=is_test,
plot_rc=plot_rc)
self.fig, self.main_ax = self.score_plot.fig, self.score_plot.main_ax
plt.subplots_adjust(bottom=0.2)
self.score_plot.draw()
max_time = self.score_plot.records.absolute_end
ax_s = plt.axes([0.15, 0.1, 0.75, 0.03])
ax_w = plt.axes([0.15, 0.05, 0.75, 0.03])
self.slider_start = Slider(ax_s, 'Start', 0., max_time, valinit=0.)
self.slider_start.on_changed(self.update)
self.slider_width = Slider(ax_w, 'Width', 0., 30., valinit=15.)
self.slider_width.on_changed(self.update)
self.fig.canvas.mpl_connect('key_press_event', self.on_key)
def update(self, val):
self.score_plot.current.absolute_start = self.slider_start.val
width = self.slider_width.val
self.score_plot.current.absolute_end = (
self.score_plot.current.absolute_start + width)
self.score_plot.draw()
def on_key(self, event):
if event.key == 'right':
direction = 1
elif event.key == 'left':
direction = -1
else:
return
self.slider_start.set_val(self.slider_start.val +
direction * .5 * self.slider_width.val)
self.update(None)
def CreateDisplay(self):
rax = plt.axes([0.025, 0.8, 0.15, 0.15])
radioSelectOperation = RadioButtons(rax, ("Search", "Insert"),
active=0)
radioSelectOperation.on_clicked(self.OnOperationTypeSelect)
radioSelectOperation.set_active(0)
axAS = plt.axes([0.25, 0.20, 0.65, 0.03])
axC = plt.axes([0.25, 0.15, 0.65, 0.03])
axGS = plt.axes([0.25, 0.10, 0.65, 0.03])
sAtomicSize = Slider(axAS,
'AtomicSize',
0,
len(self.AtomicSize) - 1,
valinit=0,
valfmt="%1.2f")
sAtomicSize.on_changed(partial(self.setAS_slider, sAtomicSize))
sAtomicSize.set_val(0.0)
sCapacity = Slider(axC,
'Capacity',
0,
len(self.Capacity) - 1,
valinit=0,
valfmt="%i")
sCapacity.on_changed(partial(self.setC_slider, sCapacity))
sCapacity.set_val(0.0)
sGridSize = Slider(axGS,
'GridSize',
0,
len(self.GridSize) - 1,
valinit=0,
valfmt="%i")
sGridSize.on_changed(partial(self.setGS_slider, sGridSize))
sGridSize.set_val(0.0)
plt.show()
class AtlasEditor(plot_support.ImageSyncMixin):
"""Graphical interface to view an atlas in multiple orthogonal
dimensions and edit atlas labels.
:attr:`plot_eds` are dictionaries of keys specified by one of
:const:`magmap.config.PLANE` plane orientations to Plot Editors.
Attributes:
image5d: Numpy image array in t,z,y,x,[c] format.
labels_img: Numpy image array in z,y,x format.
channel: Channel of the image to display.
offset: Index of plane at which to start viewing in x,y,z (user)
order.
fn_close_listener: Handle figure close events.
borders_img: Numpy image array in z,y,x,[c] format to show label
borders, such as that generated during label smoothing.
Defaults to None. If this image has a different number of
labels than that of ``labels_img``, a new colormap will
be generated.
fn_show_label_3d: Function to call to show a label in a
3D viewer. Defaults to None.
title (str): Window title; defaults to None.
fn_refresh_atlas_eds (func): Callback for refreshing other
Atlas Editors to synchronize them; defaults to None.
Typically takes one argument, this ``AtlasEditor`` object
to refreshing it. Defaults to None.
alpha_slider: Matplotlib alpha slider control.
alpha_reset_btn: Maplotlib button for resetting alpha transparency.
alpha_last: Float specifying the previous alpha value.
interp_planes: Current :class:`InterpolatePlanes` object.
interp_btn: Matplotlib button to initiate plane interpolation.
save_btn: Matplotlib button to save the atlas.
fn_status_bar (func): Function to call during status bar updates
in :class:`pixel_display.PixelDisplay`; defaults to None.
fn_update_coords (func): Handler for coordinate updates, which
takes coordinates in z-plane orientation; defaults to None.
"""
_EDIT_BTN_LBLS = ("Edit", "Editing")
def __init__(self,
image5d,
labels_img,
channel,
offset,
fn_close_listener,
borders_img=None,
fn_show_label_3d=None,
title=None,
fn_refresh_atlas_eds=None,
fig=None,
fn_status_bar=None):
"""Plot ROI as sequence of z-planes containing only the ROI itself."""
super().__init__()
self.image5d = image5d
self.labels_img = labels_img
self.channel = channel
self.offset = offset
self.fn_close_listener = fn_close_listener
self.borders_img = borders_img
self.fn_show_label_3d = fn_show_label_3d
self.title = title
self.fn_refresh_atlas_eds = fn_refresh_atlas_eds
self.fig = fig
self.fn_status_bar = fn_status_bar
self.alpha_slider = None
self.alpha_reset_btn = None
self.alpha_last = None
self.interp_planes = None
self.interp_btn = None
self.save_btn = None
self.edit_btn = None
self.color_picker_box = None
self.fn_update_coords = None
self._labels_img_sitk = None # for saving labels image
def show_atlas(self):
"""Set up the atlas display with multiple orthogonal views."""
# set up the figure
if self.fig is None:
fig = figure.Figure(self.title)
self.fig = fig
else:
fig = self.fig
fig.clear()
gs = gridspec.GridSpec(2,
1,
wspace=0.1,
hspace=0.1,
height_ratios=(20, 1),
figure=fig,
left=0.06,
right=0.94,
bottom=0.02,
top=0.98)
gs_viewers = gridspec.GridSpecFromSubplotSpec(2,
2,
subplot_spec=gs[0, 0])
# set up a colormap for the borders image if present
cmap_borders = colormaps.get_borders_colormap(self.borders_img,
self.labels_img,
config.cmap_labels)
coord = list(self.offset[::-1])
# editor controls, split into a slider sub-spec to allow greater
# spacing for labels on either side and a separate sub-spec for
# buttons and other fields
gs_controls = gridspec.GridSpecFromSubplotSpec(1,
2,
subplot_spec=gs[1, 0],
width_ratios=(1, 1),
wspace=0.15)
self.alpha_slider = Slider(
fig.add_subplot(gs_controls[0, 0]),
"Opacity",
0.0,
1.0,
valinit=plot_editor.PlotEditor.ALPHA_DEFAULT)
gs_controls_btns = gridspec.GridSpecFromSubplotSpec(
1, 5, subplot_spec=gs_controls[0, 1], wspace=0.1)
self.alpha_reset_btn = Button(fig.add_subplot(gs_controls_btns[0, 0]),
"Reset")
self.interp_btn = Button(fig.add_subplot(gs_controls_btns[0, 1]),
"Fill Label")
self.interp_planes = InterpolatePlanes(self.interp_btn)
self.interp_planes.update_btn()
self.save_btn = Button(fig.add_subplot(gs_controls_btns[0, 2]), "Save")
self.edit_btn = Button(fig.add_subplot(gs_controls_btns[0, 3]), "Edit")
self.color_picker_box = TextBox(
fig.add_subplot(gs_controls_btns[0, 4]), None)
# adjust button colors based on theme and enabled status; note
# that colors do not appear to refresh until fig mouseover
for btn in (self.alpha_reset_btn, self.edit_btn):
enable_btn(btn)
enable_btn(self.save_btn, False)
enable_btn(self.color_picker_box, color=config.widget_color + 0.1)
def setup_plot_ed(axis, gs_spec):
# set up a PlotEditor for the given axis
# subplot grid, with larger height preference for plot for
# each increased row to make sliders of approx equal size and
# align top borders of top images
rows_cols = gs_spec.get_rows_columns()
extra_rows = rows_cols[3] - rows_cols[2]
gs_plot = gridspec.GridSpecFromSubplotSpec(
2,
1,
subplot_spec=gs_spec,
height_ratios=(1, 10 + 14 * extra_rows),
hspace=0.1 / (extra_rows * 1.4 + 1))
# transform arrays to the given orthogonal direction
ax = fig.add_subplot(gs_plot[1, 0])
plot_support.hide_axes(ax)
plane = config.PLANE[axis]
arrs_3d, aspect, origin, scaling = \
plot_support.setup_images_for_plane(
plane,
(self.image5d[0], self.labels_img, self.borders_img))
img3d_tr, labels_img_tr, borders_img_tr = arrs_3d
# slider through image planes
ax_scroll = fig.add_subplot(gs_plot[0, 0])
plane_slider = Slider(ax_scroll,
plot_support.get_plane_axis(plane),
0,
len(img3d_tr) - 1,
valfmt="%d",
valinit=0,
valstep=1)
# plot editor
max_size = max_sizes[axis] if max_sizes else None
plot_ed = plot_editor.PlotEditor(
ax,
img3d_tr,
labels_img_tr,
config.cmap_labels,
plane,
aspect,
origin,
self.update_coords,
self.refresh_images,
scaling,
plane_slider,
img3d_borders=borders_img_tr,
cmap_borders=cmap_borders,
fn_show_label_3d=self.fn_show_label_3d,
interp_planes=self.interp_planes,
fn_update_intensity=self.update_color_picker,
max_size=max_size,
fn_status_bar=self.fn_status_bar)
return plot_ed
# setup plot editors for all 3 orthogonal directions
max_sizes = plot_support.get_downsample_max_sizes()
for i, gs_viewer in enumerate(
(gs_viewers[:2, 0], gs_viewers[0, 1], gs_viewers[1, 1])):
self.plot_eds[config.PLANE[i]] = setup_plot_ed(i, gs_viewer)
self.set_show_crosslines(True)
# attach listeners
fig.canvas.mpl_connect("scroll_event", self.scroll_overview)
fig.canvas.mpl_connect("key_press_event", self.on_key_press)
fig.canvas.mpl_connect("close_event", self._close)
fig.canvas.mpl_connect("axes_leave_event", self.axes_exit)
self.alpha_slider.on_changed(self.alpha_update)
self.alpha_reset_btn.on_clicked(self.alpha_reset)
self.interp_btn.on_clicked(self.interpolate)
self.save_btn.on_clicked(self.save_atlas)
self.edit_btn.on_clicked(self.toggle_edit_mode)
self.color_picker_box.on_text_change(self.color_picker_changed)
# initialize and show planes in all plot editors
if self._max_intens_proj is not None:
self.update_max_intens_proj(self._max_intens_proj)
self.update_coords(coord, config.PLANE[0])
plt.ion() # avoid the need for draw calls
def _close(self, evt):
"""Handle figure close events by calling :attr:`fn_close_listener`
with this object.
Args:
evt (:obj:`matplotlib.backend_bases.CloseEvent`): Close event.
"""
self.fn_close_listener(evt, self)
def on_key_press(self, event):
"""Respond to key press events.
"""
if event.key == "a":
# toggle between current and 0 opacity
if self.alpha_slider.val == 0:
# return to saved alpha if available and reset
if self.alpha_last is not None:
self.alpha_slider.set_val(self.alpha_last)
self.alpha_last = None
else:
# make translucent, saving alpha if not already saved
# during a halve-opacity event
if self.alpha_last is None:
self.alpha_last = self.alpha_slider.val
self.alpha_slider.set_val(0)
elif event.key == "A":
# halve opacity, only saving alpha on first halving to allow
# further halving or manual movements while still returning to
# originally saved alpha
if self.alpha_last is None:
self.alpha_last = self.alpha_slider.val
self.alpha_slider.set_val(self.alpha_slider.val / 2)
elif event.key == "up" or event.key == "down":
# up/down arrow for scrolling planes
self.scroll_overview(event)
elif event.key == "w":
# shortcut to toggle editing mode
self.toggle_edit_mode(event)
elif event.key == "ctrl+s" or event.key == "cmd+s":
# support default save shortcuts on multiple platforms;
# ctrl-s will bring up save dialog from fig, but cmd/win-S
# will bypass
self.save_fig(self.get_save_path())
def update_coords(self, coord, plane_src=config.PLANE[0]):
"""Update all plot editors with given coordinates.
Args:
coord: Coordinate at which to center images, in z,y,x order.
plane_src: One of :const:`magmap.config.PLANE` to specify the
orientation from which the coordinates were given; defaults
to the first element of :const:`magmap.config.PLANE`.
"""
coord_rev = libmag.transpose_1d_rev(list(coord), plane_src)
for i, plane in enumerate(config.PLANE):
coord_transposed = libmag.transpose_1d(list(coord_rev), plane)
if i == 0:
self.offset = coord_transposed[::-1]
if self.fn_update_coords:
# update offset based on xy plane, without centering
# planes are centered on the offset as-is
self.fn_update_coords(coord_transposed, False)
self.plot_eds[plane].update_coord(coord_transposed)
def view_subimg(self, offset, shape):
"""Zoom all Plot Editors to the given sub-image.
Args:
offset: Sub-image coordinates in ``z,y,x`` order.
shape: Sub-image shape in ``z,y,x`` order.
"""
for i, plane in enumerate(config.PLANE):
offset_tr = libmag.transpose_1d(list(offset), plane)
shape_tr = libmag.transpose_1d(list(shape), plane)
self.plot_eds[plane].view_subimg(offset_tr[1:], shape_tr[1:])
self.fig.canvas.draw_idle()
def refresh_images(self, plot_ed=None, update_atlas_eds=False):
"""Refresh images in a plot editor, such as after editing one
editor and updating the displayed image in the other editors.
Args:
plot_ed (:obj:`magmap.plot_editor.PlotEditor`): Editor that
does not need updating, typically the editor that originally
changed. Defaults to None.
update_atlas_eds (bool): True to update other ``AtlasEditor``s;
defaults to False.
"""
for key in self.plot_eds:
ed = self.plot_eds[key]
if ed != plot_ed: ed.refresh_img3d_labels()
if ed.edited:
# display save button as enabled if any editor has been edited
enable_btn(self.save_btn)
if update_atlas_eds and self.fn_refresh_atlas_eds is not None:
# callback to synchronize other Atlas Editors
self.fn_refresh_atlas_eds(self)
def scroll_overview(self, event):
"""Scroll images and crosshairs in all plot editors
Args:
event: Scroll event.
"""
for key in self.plot_eds:
self.plot_eds[key].scroll_overview(event)
def alpha_update(self, event):
"""Update the alpha transparency in all plot editors.
Args:
event: Slider event.
"""
for key in self.plot_eds:
self.plot_eds[key].alpha_updater(event)
def alpha_reset(self, event):
"""Reset the alpha transparency in all plot editors.
Args:
event: Button event, currently ignored.
"""
self.alpha_slider.reset()
def axes_exit(self, event):
"""Trigger axes exit for all plot editors.
Args:
event: Axes exit event.
"""
for key in self.plot_eds:
self.plot_eds[key].on_axes_exit(event)
def interpolate(self, event):
"""Interpolate planes using :attr:`interp_planes`.
Args:
event: Button event, currently ignored.
"""
try:
self.interp_planes.interpolate(self.labels_img)
# flag Plot Editors as edited so labels can be saved
for ed in self.plot_eds.values():
ed.edited = True
self.refresh_images(None, True)
except ValueError as e:
print(e)
def save_atlas(self, event):
"""Save atlas labels using the registered image suffix given by
:attr:`config.reg_suffixes[config.RegSuffixes.ANNOTATION]`.
Args:
event: Button event, currently not used.
"""
# only save if at least one editor has been edited
if not any([ed.edited for ed in self.plot_eds.values()]): return
# save to the labels reg suffix; use sitk Image if loaded and store
# any Image loaded during saving
reg_name = config.reg_suffixes[config.RegSuffixes.ANNOTATION]
if self._labels_img_sitk is None:
self._labels_img_sitk = config.labels_img_sitk
self._labels_img_sitk = sitk_io.write_registered_image(
self.labels_img,
config.filename,
reg_name,
self._labels_img_sitk,
overwrite=True)
# reset edited flag in all editors and show save button as disabled
for ed in self.plot_eds.values():
ed.edited = False
enable_btn(self.save_btn, False)
print("Saved labels image at {}".format(datetime.datetime.now()))
def get_save_path(self):
"""Get figure save path based on filename, ROI, and overview plane
shown.
Returns:
str: Figure save path.
"""
ext = config.savefig if config.savefig else config.DEFAULT_SAVEFIG
return "{}.{}".format(
naming.get_roi_path(os.path.basename(self.title), self.offset),
ext)
def toggle_edit_mode(self, event):
"""Toggle editing mode, determining the current state from the
first :class:`magmap.plot_editor.PlotEditor` and switching to the
opposite value for all plot editors.
Args:
event: Button event, currently not used.
"""
edit_mode = False
for i, ed in enumerate(self.plot_eds.values()):
if i == 0:
# change edit mode based on current mode in first plot editor
edit_mode = not ed.edit_mode
toggle_btn(self.edit_btn, edit_mode, text=self._EDIT_BTN_LBLS)
ed.edit_mode = edit_mode
if not edit_mode:
# reset the color picker text box when turning off editing
self.color_picker_box.set_val("")
def update_color_picker(self, val):
"""Update the color picker :class:`TextBox` with the given value.
Args:
val (str): Color value. If None, only :meth:`color_picker_changed`
will be triggered.
"""
if val is None:
# updated picked color directly
self.color_picker_changed(val)
else:
# update text box, which triggers color_picker_changed
self.color_picker_box.set_val(val)
def color_picker_changed(self, text):
"""Respond to color picker :class:`TextBox` changes by updating
the specified intensity value in all plot editors.
Args:
text (str): String of text box value. Converted to an int if
non-empty.
"""
intensity = text
if text:
if not libmag.is_number(intensity): return
intensity = int(intensity)
print("updating specified color to", intensity)
for i, ed in enumerate(self.plot_eds.values()):
ed.intensity_spec = intensity
class VideoViewer(object):
"""
A matplotlib-based video viewer.
Parameters
----------
video : list-like, iterator
A list of a tuple of 2D arrays or a generator of a tuple of 2D arrays.
If an iterator is provided, you must set 'count' as well.
count: int
Length of the video. When this is set it displays only first 'count' frames of the video.
id : int, optional
For multi-frame data specifies camera index.
norm_func : callable
Normalization function that takes a single argument (array) and returns
a single element (array). Can be used to apply custom normalization
function to the image before it is shown.
title : str, optional
Plot title.
kw : options, optional
Extra arguments passed directly to imshow function
Examples
--------
>>> from cddm.viewer import VideoViewer
>>> video = (np.random.randn(256,256) for i in range(256))
>>> vg = VideoViewer(video, 256, title = "iterator example") #must set nframes, because video has no __len__
#>>> vg.show()
>>> video = [np.random.randn(256,256) for i in range(256)]
>>> vl = VideoViewer(video, title = "list example")
#>>> vl.show()
"""
def __init__(self,
video,
count=None,
id=0,
norm_func=lambda x: x.real,
title="",
**kw):
if count is None:
try:
count = len(video)
except TypeError:
raise Exception("You must specify count!")
self._norm = norm_func
self.id = id
self.index = 0
self.video = video
self.fig, self.ax = plt.subplots()
self.ax.set_title(title)
plt.subplots_adjust(bottom=0.25)
frame = next(iter(video)) #take first frame
frame = self._prepare_image(frame)
self.img = self.ax.imshow(frame, **kw)
self.fig.colorbar(self.img, ax=self.ax)
self.axframe = plt.axes([0.1, 0.1, 0.7, 0.03])
self.sframe = Slider(self.axframe,
'',
0,
count - 1,
valinit=0,
valstep=1,
valfmt='%i')
self.axnext = plt.axes([0.7, 0.02, 0.1, 0.05])
self.bnext = Button(self.axnext, '>')
self.axnext2 = plt.axes([0.6, 0.02, 0.1, 0.05])
self.bnext2 = Button(self.axnext2, '>>>')
self.axprev = plt.axes([0.1, 0.02, 0.1, 0.05])
self.bprev = Button(self.axprev, '<')
self.axprev2 = plt.axes([0.2, 0.02, 0.1, 0.05])
self.bprev2 = Button(self.axprev2, '<<<')
self.axstop = plt.axes([0.4, 0.02, 0.1, 0.05])
self.bstop = Button(self.axstop, 'Stop')
self.axplay = plt.axes([0.3, 0.02, 0.1, 0.05])
self.bplay = Button(self.axplay, 'Play')
self.axfast = plt.axes([0.5, 0.02, 0.1, 0.05])
self.bfast = Button(self.axfast, 'FF')
self.playing = False
self.step_fast = count / 100
self.step = 1
self.pause_duration = 0.001
def _play():
while self.playing:
plt.pause(self.pause_duration)
next_frame = self.sframe.val + self.step
if next_frame >= count:
self.playing = False
else:
self.sframe.set_val(next_frame)
def stop(event):
self.playing = False
@skip_runtime_error
def play(event):
self.playing = True
self.step = 1
_play()
@skip_runtime_error
def play_fast(event):
self.playing = True
self.step = self.step_fast
_play()
def next_frame(event):
self.sframe.set_val(min(self.sframe.val + 1, count - 1))
def next_fast(event):
self.sframe.set_val(min(self.sframe.val + self.step, count - 1))
def prev_frame(event):
self.sframe.set_val(max(self.sframe.val - 1, 0))
def prev_fast(event):
self.sframe.set_val(max(self.sframe.val - self.step, 0))
@skip_runtime_error
def update(val):
i = int(self.sframe.val)
try:
frame = self.video[i] #assume list-like object
self.index = i
except TypeError:
#assume generator
frame = None
if i > self.index:
for frame in self.video:
self.index += 1
if self.index >= i:
break
if frame is not None:
frame = self._prepare_image(frame)
self.img.set_data(frame)
self.fig.canvas.draw_idle()
self.sframe.on_changed(update)
self.bnext.on_clicked(next_frame)
self.bstop.on_clicked(stop)
self.bplay.on_clicked(play)
self.bfast.on_clicked(play_fast)
self.bnext2.on_clicked(next_fast)
self.bprev2.on_clicked(prev_fast)
self.bprev.on_clicked(prev_frame)
def _prepare_image(self, im):
if isinstance(im, tuple) or isinstance(im, list):
return self._norm(im[self.id])
else:
return self._norm(im)
def show(self):
"""Shows video."""
plt.show()
class Control:
def __init__(self, figure, position, label, initial_value, dim):
"""
Control group for slice
"""
# Slider
self.slider = Slider(figure.add_axes(position,
xticks=[],
yticks=[],
facecolor='#222222'),
label,
0,
dim - 1,
valinit=initial_value,
valstep=1,
valfmt='%1.0f',
color='#444444')
# Set Button Positions
position[1] = position[1] - 0.04
position[2] = position[2] / 2
position[3] = 0.04
self.buttondown = Button(figure.add_axes(position),
'-',
color='#222222',
hovercolor='#333333')
# Buttons
position[0] = position[0] + position[2]
self.buttonup = Button(figure.add_axes(position),
'+',
color='#222222',
hovercolor='#333333')
# save value for display
self.value = initial_value
# save dim for slider limit
self.lim = dim
def get_value(self):
"""
Returns the current value of the control
"""
return int(self.value)
# decrement slider
def decrement(self, event):
new_val = self.value - 1
if new_val >= 0:
self.slider.set_val(new_val)
# increment slider
def increment(self, event):
new_val = self.value + 1
if new_val < self.lim:
self.slider.set_val(new_val)
# update control
def update(self, value, callback_func):
"""
Updates the current value of the control then runs the
callback_func with the current value
"""
self.value = int(value) # set new value
callback_func(int(value)) # execute callback
class ytViewer(object):
def __init__(self, filename, fold=19277, nmax=100,NORM=True,dtype=int8,DEB=0,shear_val=0.):
self.UPDATE = True
self.color = True
self.NORM = NORM
self.NMAX = nmax
self.fold = fold
self.increment = 5
self.index = 0
self.shear_val = shear_val
self.remove_len1 = 0
self.remove_len2 = 1
self.fig = figure(figsize=(16,7))
self.data = fromfile(filename,dtype=dtype)
self.max_index = int(len(self.data)/self.fold)
self.data = self.data[:self.max_index*self.fold]
self.folded_data_orig2 = self.data.reshape(self.max_index,self.fold)
self.folded_data_orig = array(self.folded_data_orig2)
self.folded_data_orig3 = array(self.folded_data_orig)
self.folded_data = self.folded_data_orig3[:self.NMAX]
self.Y0 = 0
self.ax = axes([0.1,0.4,0.8,0.47])
if not self.NORM:
self.im = self.ax.imshow(self.folded_data, interpolation='nearest', aspect='auto', origin='lower', vmin=0, vmax=255)
else:
self.im = self.ax.imshow(self.folded_data, interpolation='nearest', aspect='auto', origin='lower', vmin=self.data.min(), vmax=self.data.max())
self.cursor = Cursor(self.ax, useblit=True, color='red', linewidth=2)
self.axh = axes([0.1,0.05,0.8,0.2])
self.hline, = self.axh.plot(self.folded_data[self.Y0,:])
self.axh.set_xlim(0,len(self.folded_data[0,:]))
self.axh.set_ylim(self.folded_data.min(),self.folded_data.max())
# create 'remove_len1' slider
self.remove_len1_sliderax = axes([0.1,0.925,0.8,0.02])
self.remove_len1_slider = Slider(self.remove_len1_sliderax,'beg',0.,self.fold*(3.5/4),self.remove_len1,'%d')
self.remove_len1_slider.on_changed(self.update_tab)
# create 'remove_len2' slider
self.remove_len2_sliderax = axes([0.1,0.905,0.8,0.02])
self.remove_len2_slider = Slider(self.remove_len2_sliderax,'end',0.,self.fold*(3.5/4),self.remove_len2,'%d')
self.remove_len2_slider.on_changed(self.update_tab)
# create 'shear' slider
self.shear_sliderax = axes([0.1,0.88,0.8,0.02])
self.shear_slider = Slider(self.shear_sliderax,'shear',-0.5,0.5,self.shear_val,'%1.3f')
self.shear_slider.on_changed(self.update_shear)
# create 'index' slider
self.index_sliderax = axes([0.1,0.975,0.8,0.02])
self.index_slider = Slider(self.index_sliderax,'index',0,self.max_index-self.increment,0,'%d')
self.index_slider.on_changed(self.update_param)
# create 'nmax' slider
self.nmax_sliderax = axes([0.1,0.955,0.8,0.02])
self.nmax_slider = Slider(self.nmax_sliderax,'nmax',0,self.max_index,self.NMAX,'%d')
self.nmax_slider.on_changed(self.update_tab)
cid = self.fig.canvas.mpl_connect('motion_notify_event', self.mousemove)
cid2 = self.fig.canvas.mpl_connect('key_press_event', self.keypress)
self.axe_toggledisplay = self.fig.add_axes([0.43,0.27,0.14,0.1])
self.plot_circle(0,0,2,fc='#00FF7F')
mpl.pyplot.axis('off')
if self.shear_val!=0.: self.shear()
gobject.idle_add(self.update_plot)
show()
def update_shear(self,value):
self.shear_val = round(self.shear_slider.val,3)
self.shear()
self.update_tab(value)
def update_param(self,value):
self.index = int(round(self.index_slider.val,0))
self.update_tab(value)
def shear(self):
if self.shear_val == 0:
pass
dd = array(self.folded_data_orig2)
for i in range(0,self.folded_data_orig2.shape[0]):
dd[i,:] = roll(self.folded_data_orig2[i,:], int(i*self.shear_val))
self.folded_data_orig = dd
def update_tab(self,val):
self.remove_len1 = int(self.remove_len1_slider.val)
self.remove_len2 = int(self.remove_len2_slider.val)
self.NMAX = int(round(self.nmax_slider.val,0))
self.folded_data_orig3 = array(self.folded_data_orig[:,self.remove_len1:-self.remove_len2])
self.folded_data = self.folded_data_orig3[self.index:(self.index+self.NMAX)]
self.Y0 = 0
self.ax.clear()
self.im = self.ax.imshow(self.folded_data, interpolation='nearest', aspect='auto', origin='lower', vmin=self.data.min(), vmax=self.data.max())
self.axh = axes([0.1,0.05,0.8,0.2])
self.axh.clear()
self.hline, = self.axh.plot(self.folded_data[self.Y0,:])
plt.ylim(self.folded_data.min(),self.folded_data.max())
plt.xlim(0,len(self.folded_data[self.Y0,:]))
draw()
def update_plot(self):
while self.UPDATE:
self.folded_data = self.folded_data_orig3[self.index:(self.index+self.NMAX)]
### Update picture ###
self.im.set_data(self.folded_data)
self.hline.set_ydata(self.folded_data[self.Y0,:])
self.index = self.index + self.increment
self.index_slider.set_val(self.index)
draw()
return True
return False
def update_cut(self):
self.hline.set_ydata(self.folded_data[self.Y0,:])
draw()
def keypress(self, event):
if event.key == 'q': # eXit
del event
sys.exit()
elif event.key=='n':
del event
self.NORM = not(self.NORM)
if not self.NORM:
self.ax.clear()
self.im = self.ax.imshow(self.folded_data, interpolation='nearest', aspect='auto',
origin='lower', vmin=0, vmax=255)
self.axh.set_ylim(0, 255)
else:
self.ax.clear()
self.im = self.ax.imshow(self.folded_data, interpolation='nearest', aspect='auto',
origin='lower', vmin=self.folded_data.min(), vmax=self.folded_data.max())
self.axh.set_ylim(self.folded_data.min(), self.folded_data.max())
elif event.key == ' ': # play/pause
self.toggle_update()
else:
print 'Key '+str(event.key)+' not known'
def mousemove(self, event):
# called on each mouse motion to get mouse position
if event.inaxes!=self.ax: return
self.X0 = int(round(event.xdata,0))
self.Y0 = int(round(event.ydata,0))
self.update_cut()
def toggle_update(self):
self.UPDATE = not(self.UPDATE)
if self.UPDATE:
gobject.idle_add(self.update_plot)
self.color = not(self.color)
if not(self.color):
self.patch.remove()
self.axe_toggledisplay = self.fig.add_axes([0.43,0.27,0.14,0.1])
self.axe_toggledisplay.clear()
self.plot_circle(0,0,2,fc='#FF4500')
mpl.pyplot.axis('off')
draw()
else:
self.patch.remove()
self.axe_toggledisplay = self.fig.add_axes([0.43,0.27,0.14,0.1])
self.axe_toggledisplay.clear()
self.plot_circle(0,0,2,fc='#00FF7F')
mpl.pyplot.axis('off')
draw()
#gobject.idle_add(self.update_plot)
def plot_circle(self,x,y,r,fc='r'):
"""Plot a circle of radius r at position x,y"""
cir = mpl.patches.Circle((x,y), radius=r, fc=fc)
self.patch = mpl.pyplot.gca().add_patch(cir)
class Notes(object):
"""
An interactive matplotlib window for ROI drawing, definition of the main
axis of the eye, and selection of a reference frames where the eye is open.
"""
def __init__(self, stack, frame_lut, key):
self.wintitle = "ixtract - " + key['m']
# stack of frames sampled from the files to be analyzed:
self.stack = stack
self.nframes = self.stack.shape[0]
self.frame_lut = frame_lut # look-up table for frame indices
self.ref_stack = [] # selected frames
self.ref_frame_inds = [] # indices of selected frames
self.mode = 'roi' # init in roi drawing mode
self.points = np.zeros((2,2)) # holds mouse click coords
self.roi = None # init ROI and axis variables to None-type
self.axis = None
self.line = None # patch variables to display ROI and axis on figure
self.rect = None
# set up frame display axis
self.fig, _ = plt.subplots()
self.fig.canvas.set_window_title(self.wintitle)
grid = gs.GridSpec(9, 12)
self.ax_frame = plt.subplot(grid[:8,:])
self.ax_frame.axis('off')
self.ax_frame.imshow(self.stack[0,:,:], cmap='gray')
# set up slider axis
self.ax_slider = plt.subplot(grid[8,0:11])
self.slider = Slider(self.ax_slider, 'Frame', 0, self.nframes-1,
valinit=0, valfmt='%d')
# connect callback functions
# REMOVE AXIS BINDINGS (AXIS CHECK IS IN CALLBACK FUNCTIONS)
self.cidpress = self.ax_frame.figure.canvas.mpl_connect(
'button_press_event', self.on_click)
self.cidrelease = self.ax_frame.figure.canvas.mpl_connect(
'button_release_event', self.on_release)
self.slider.on_changed(self.update_frame)
# disable default matplotlib key bindings
manager, canvas = self.fig.canvas.manager, self.fig.canvas
canvas.mpl_disconnect(manager.key_press_handler_id)
# connect to custom key bindings
self.cidkey = self.fig.canvas.mpl_connect('key_press_event', self.on_key)
# display user prompts
print("Select an ROI and define the main axis of the eye")
print(" - use 'd' to switch between drawing modes")
print(" - use 's' to select a few reference frames in which the eye is open")
print(" - click on the slider or use the arrow keys to scroll between video frames")
print(" - use 'c' to continue, or 'esc' to quit")
def on_click(self, event):
"""stores coordintes of click"""
if event.inaxes != self.ax_frame:
return
self.points[0,0], self.points[0,1] = event.xdata, event.ydata
def on_release(self, event):
"""assigns coordinates of click & release to ROI or axis properties
and draws the appropriate shape patch onto the figure"""
if event.inaxes != self.ax_frame:
return
# assign click data to temporary variables
self.points[1,0], self.points[1,1] = event.xdata, event.ydata
# assign click data as object property and draw patch
if self.mode == 'roi':
# parse click data
self.roi = self.points.copy()
x0 = min(self.points[:,0])
y0 = min(self.points[:,1])
dx = np.absolute(self.points[1,0] - self.points[0,0]) # width
dy = np.absolute(self.points[1,1] - self.points[0,1]) # height
# remove old patch
if self.rect is not None:
self.rect.remove()
# add new patch
self.rect = Rectangle((x0,y0), dx, dy, lw=2, ec=[0,1,0.5], fill=False)
self.ax_frame.add_patch(self.rect)
elif self.mode == 'axis':
# parse click data
self.axis = self.points.copy()
x0 = self.points[0,0]
y0 = self.points[0,1]
dx = self.points[1,0] - self.points[0,0]
dy = self.points[1,1] - self.points[0,1]
# remove old patch
if self.line is not None:
self.line.remove()
# add new patch
# arrow patch used in lieu of a line patch (arrow head width set to zero)
self.line = Arrow(x0, y0, dx, dy, width=0, lw=2, color=[1,0,1])
self.ax_frame.add_patch(self.line)
# update figure
self.fig.canvas.draw()
def on_key(self, event):
"""specifies functions of pressed keys in matplotlib figure"""
# change drawing mode
if event.key == 'd':
if self.mode == 'roi':
self.mode = 'axis'
elif self.mode == 'axis':
self.mode = 'roi'
print(" Drawing: " + self.mode)
# select current frame as open-eye reference
elif event.key == 's':
newframe = np.int(np.round(self.slider.val))
print(" Frame %d selected as reference" % newframe)
self.ref_stack.append(self.stack[newframe,:,:])
self.ref_frame_inds.append(newframe)
# change frame
elif event.key == 'right': # move forward one frame
if np.round(self.slider.val) < self.nframes-1:
self.slider.set_val(self.slider.val+1)
elif event.key == 'left': # move back one frame
if np.round(self.slider.val) > 0:
self.slider.set_val(self.slider.val-1)
# confirm or exit
elif event.key == 'c': # confirm and continue
# check that all annotations have been made
if self.roi is None:
print(" Please define an ROI")
elif self.axis is None:
print(" Please define the eye's aixs")
elif len(self.ref_stack) == 0:
print(" Please select at least one open-eye reference frame")
else:
print("ROI and axis confirmed")
self.wrap_up()
plt.close('all')
elif event.key == 'escape': # quit eye tracking
print("Eye tracking aborted")
plt.close('all')
sys.exit()
def update_frame(self, val):
"""Updates frame based on slider value"""
newframe = np.int(np.round(self.slider.val))
frame = self.stack[newframe,:,:]
self.ax_frame.imshow(frame, cmap='gray')
def wrap_up(self):
"""Performs cropping and reference frame computations before closing"""
# crop stacks based on selected ROI
print(" Cropping frames...")
self.stack = img.crop(self.stack, self.roi)
self.ref_stack = img.crop(np.array(self.ref_stack), self.roi)
# take median if multiple frames are given as reference
if len(self.ref_stack.shape) == 2:
self.ref_frame = self.ref_stack
elif len(self.ref_stack.shape) == 3:
self.ref_frame = np.median(self.ref_stack, axis=0)
# compute mean pixel-wise differences from the reference frame
print(" Comparing frames to reference...")
self.diffs = img.mean_diffs(self.stack, self.ref_frame)
# store indices of frames used as reference
self.ref_frames = self.frame_lut[self.ref_frame_inds]
def pager(L,shape,vmake,vpaint,offset=0,save=None,savedefaults=dict(dest='pager-sav',format='svg'),bstyle={},**_ka):
r"""
:param L: a list of arbitrary objects other than :const:`None`
:param shape: a pair (number of rows, number of columns) or a single number if they are equat
:param vmake: a function to initialise the display (see below)
:param vpaint: a function to instantiate the display for a given page (see below)
:param savedefaults: used as default keyword arguments of method :meth:`savefig` when saving pages
:type savedefaults: :class:`dict`
This function first create a :class:`Cell` instance with all the remaining arguments, then splits it into a grid of sub-cells according to *shape*, then displays *L* page per page on the grid. Each page displays a slice of *L* of length equal to the product of the components of *shape* (or less, for the final page). The toolbar is enriched with page navigation buttons. A save button also allows to save the whole collection of pages in a given directory (beware: may be long).
Function *vmake* takes as input a :class:`Cell` instance and instantiates it as needed. It can store information (e.g. about the specific role of each artist created in the cell), if needed, by simply setting attributes in the cell. This is called once for each cell at the begining of the display.
Function *vpaint* takes as input a cell and an element of *L* or None, and displays that element in the cell (or resets the cell to indicate a missing value), possibly using the artists created by *vmake* and stored in the cell. This is called once at each page display and for each cell.
Unfortunately, matplotlib toolbars are not standardised: the depend on the backend and may not support adding button.
"""
#------------------------------------------------------------------------------
from numpy import ceil, rint, clip
from matplotlib.text import Text
from matplotlib.widgets import Slider
from matplotlib.pyplot import close
from pathlib import Path
from shutil import rmtree
def gen(L):
yield from L
while True: yield None
def genc(cell):
Nr,Nc = cell.shape
yield from (cell[row,col] for row in range(Nr) for col in range(Nc))
def paintp(cell,p,draw=True):
for c,x in zip(genc(cell),gen(L[p*cellpp:])): vpaint(c,x)
if draw: cell.figure.canvas.draw()
def toggle_ctrl():
ctrl.ax.set_visible(not ctrl.ax.get_visible())
cell.figure.canvas.draw()
def save_all():
ka = _ka.copy()
ka.update(fig=None,figsize=((cell.figure.get_figwidth(),cell.figure.get_figheight())))
#import multiprocessing
#multiprocessing.get_context('spawn').Process(target=pager,args=(L,shape,vmake,vpaint),kwargs=dict(save={},savedefaults=savedefaults,**ka)).start()
pager(L,shape,vmake,vpaint,save={},savedefaults=savedefaults,**ka)
cell = Cell.create(**_ka)
Nr,Nc = (shape,shape) if isinstance(shape,int) else shape
cell.make_grid(Nr,Nc)
for c in genc(cell): vmake(c)
cellpp = Nr*Nc
npage = int(ceil(len(L)/cellpp))
if save is None:
actions = [
('<<',(lambda:ctrl.set_val(clip(ctrl.val-1,1,npage)))),
('>>',(lambda:ctrl.set_val(clip(ctrl.val+1,1,npage)))),
('toggle-ctrl',toggle_ctrl),
('save-all',save_all),
]
try: menu = cell.figure.canvas.toolbar; menu.addAction
except: menu = Menu(cell.figure,**bstyle)
for a,f in actions: menu.addAction(a,f)
ctrl = Slider(cell.figure.add_axes((0.1,0.,.8,.03),visible=False,zorder=1),'page',.5,npage+.5,valinit=0,valfmt='%.0f/{}'.format(npage),closedmin=False,closedmax=False)
ctrl.on_changed(lambda p:paintp(cell,int(rint(p))-1))
ctrl.set_val(1+offset/cellpp)
else:
s = savedefaults.copy()
s.update(save)
pth = Path(s.pop('dest'))
try:
assert pth.is_dir()
for f in list(pth.iterdir()): rmtree(str(f))
except Exception as e:
logger.warn('Error on save directory %s: %s',path,e)
raise
try:
for p in range(npage):
paintp(cell,p,False)
cell.figure.savefig(str((pth/'p{:02d}'.format(p)).with_suffix('.'+s['format'])),**s)
except Exception as e: logger.warn('Error saving page %s: %s',p,e)
close(cell.figure.number)
class InteractiveView:
def __init__(self, img, peaks):
import matplotlib.pyplot as plt
from matplotlib.widgets import Button, Slider
self.fig = plt.figure(figsize=(15, 10))
self.ax = self.fig.add_subplot(111)
plt.subplots_adjust(bottom=0.15)
self.i = 1
self.peaks = peaks
self.img = img
if self.img.ndim > 2:
flatted_dim = [reduce(operator.mul, self.img.shape[:-2])]
self.img.shape = flatted_dim + list(self.img.shape[-2:])
self.text = plt.figtext(0.06, 0.05, '', transform=self.fig.transFigure)
w = 0.1
h = 0.050
y_pos = 0.04
self.axprev = plt.axes([0.7, y_pos, w, h])
self.bprev = Button(self.axprev, 'Previous')
self.bprev.on_clicked(self.prev)
self.axnext = plt.axes([0.8, y_pos, w, h])
self.bnext = Button(self.axnext, 'Next')
self.bnext.on_clicked(self.next)
self.axslide = plt.axes([0.4, 0.04, 0.25, 0.03])
self.slider = Slider(self.axslide, 'Frames', 1, int(len(self.img)),
valinit=1)
self.slider.on_changed(self.slide)
self.artists = []
self.im = None
self.draw(0)
def draw(self, i):
from matplotlib.patches import Circle
import matplotlib.pyplot as plt
self.i = i
if i > len(self.img) or i <= 0:
self.i = 1
i = 1
if self.im:
self.im.remove()
for art in self.artists:
art.remove()
self.artists = []
try:
current_peaks = self.peaks.ix[i-1]
for j, data in current_peaks.iterrows():
x = data['x']
y = data['y']
w = data['w']
outline = Circle((y, x), w, alpha=0.4, color='red')
pt = self.ax.add_patch(outline)
self.artists.append(pt)
pt = self.ax.scatter(y, x, marker='+')
self.artists.append(pt)
n_peaks = current_peaks.shape[0]
except:
n_peaks = 0
self.text.set_text("Frame %i/%i | Peaks number %i" % (i,
len(self.img),
n_peaks))
self.im = self.ax.imshow(self.img[i-1], interpolation='none', cmap='gray', shape=(2, 2))
plt.draw()
def next(self, event=None):
self.slider.set_val(self.i + 1)
def prev(self, event=None):
self.slider.set_val(self.i - 1)
def slide(self, event):
self.draw(int(event))
def show(self):
self.fig.show()
class SelectFromCollection(object):
"""Interactive RLS classifier interface for image segmentation
Parameters
----------
fig : matplotlib.figure.Figure
The Figure object on which the interface is drawn.
mmc : rlscore.learner.interactive_rls_classifier.InteractiveRlsClassifier
Interactive RLS classifier object
img : numpy.array
Array consisting of image data
collection : numpy.array, shape = [n_pixels, 2]
array consisting of the (x,y) coordinates of all usable pixels in the image
windowsize : int
Determines the size of a window around grid points (2 * windowsize + 1)
"""
def __init__(self, fig, mmc, img, collection, windowsize = 0):
#Initialize the main axis
ax = fig.add_axes([0.1,0.1,0.8,0.8])
ax.set_yticklabels([])
ax.yaxis.set_tick_params(size = 0)
ax.set_xticklabels([])
ax.xaxis.set_tick_params(size = 0)
self.imdata = ax.imshow(img)
#Initialize LassoSelector on the main axis
self.lasso = LassoSelector(ax, onselect = self.onselect)
self.lasso.connect_event('key_press_event', self.onkeypressed)
self.lasso.line.set_visible(False)
self.mmc = mmc
self.img = img
self.img_orig = img.copy()
self.collection = collection
self.selectedset = set([])
self.lockedset = set([])
self.windowsize = windowsize
#Initialize the fraction slider
self.slider_axis = fig.add_axes([0.2, 0.06, 0.6, 0.02])
self.in_selection_slider = Slider(self.slider_axis,
'Fraction slider',
0.,
1,
valinit = len(np.nonzero(self.mmc.classvec_ws)[0]) / len(mmc.working_set))
def sliderupdate(val):
val = int(val * len(mmc.working_set))
nonzeroc = len(np.nonzero(self.mmc.classvec_ws)[0])
if val > nonzeroc:
claims = val - nonzeroc
newclazz = 1
elif val < nonzeroc:
claims = nonzeroc - val
newclazz = 0
else: return
print('Claimed', claims, 'points for class', newclazz)
self.claims = claims
mmc.claim_n_points(claims, newclazz)
self.redrawall()
self.in_selection_slider.on_changed(sliderupdate)
#Initialize the display for the RLS objective funtion
self.objfun_display_axis = fig.add_axes([0.1, 0.96, 0.8, 0.02])
self.objfun_display_axis.imshow(mmc.compute_steepness_vector()[np.newaxis, :], cmap = plt.get_cmap("Oranges"))
self.objfun_display_axis.set_aspect('auto')
self.objfun_display_axis.set_yticklabels([])
self.objfun_display_axis.yaxis.set_tick_params(size = 0)
def onselect(self, verts):
#Select a new working set
self.path = Path(verts)
self.selectedset = set(np.nonzero(self.path.contains_points(self.collection))[0])
print('Selected ' + str(len(self.selectedset)) + ' points')
newws = list(self.selectedset - self.lockedset)
self.mmc.new_working_set(newws)
self.redrawall()
def onkeypressed(self, event):
print('You pressed', event.key)
if event.key == '1':
print('Assigned all selected points to class 1')
newclazz = 1
mmc.claim_all_points_in_working_set(newclazz)
if event.key == '0':
print('Assigned all selected points to class 0')
newclazz = 0
mmc.claim_all_points_in_working_set(newclazz)
if event.key == 'a':
print('Selected all points')
newws = list(set(range(len(self.collection))) - self.lockedset)
self.mmc.new_working_set(newws)
self.lasso.line.set_visible(False)
if event.key == 'c':
changecount = mmc.cyclic_descent_in_working_set()
print('Performed ', changecount, 'cyclic descent steps')
if event.key == 'l':
print('Locked the class labels of selected points')
self.lockedset = self.lockedset | self.selectedset
newws = list(self.selectedset - self.lockedset)
self.mmc.new_working_set(newws)
if event.key == 'u':
print('Unlocked the selected points')
self.lockedset = self.lockedset - self.selectedset
newws = list(self.selectedset - self.lockedset)
self.mmc.new_working_set(newws)
if event.key == 'p':
print('Compute predictions and AUC on data')
preds = self.mmc.predict(Xmat)
print(auc(mmc.Y[:, 0], preds[:, 0]))
self.redrawall()
def redrawall(self):
#Color all class one labeled pixels red
oneclazz = np.nonzero(self.mmc.classvec)[0]
col_row = self.collection[oneclazz]
rowcs, colcs = col_row[:, 1], col_row[:, 0]
red = np.array([255, 0, 0])
for i in range(-self.windowsize, self.windowsize + 1):
for j in range(-self.windowsize, self.windowsize + 1):
self.img[rowcs+i, colcs+j, :] = red
#Return the original color of the class zero labeled pixels
zeroclazz = np.nonzero(self.mmc.classvec - 1)[0]
col_row = self.collection[zeroclazz]
rowcs, colcs = col_row[:, 1], col_row[:, 0]
for i in range(-self.windowsize, self.windowsize + 1):
for j in range(-self.windowsize, self.windowsize + 1):
self.img[rowcs+i, colcs+j, :] = self.img_orig[rowcs+i, colcs+j, :]
self.imdata.set_data(self.img)
#Update the slider position according to labeling of the current working set
sliderval = 0
if len(mmc.working_set) > 0:
sliderval = len(np.nonzero(self.mmc.classvec_ws)[0]) / len(mmc.working_set)
self.in_selection_slider.set_val(sliderval)
#Update the RLS objective function display
self.objfun_display_axis.imshow(mmc.compute_steepness_vector()[np.newaxis, :], cmap=plt.get_cmap("Oranges"))
self.objfun_display_axis.set_aspect('auto')
#Final stuff
self.lasso.canvas.draw_idle()
plt.draw()
print_instructions()
class FindCenters(pb.PointBrowser):
"""
Semi-automatic fitting of bright points in TEM image
dragging ... (opt) if True, dragging is allowed for sliders
"""
def __init__(self, image, dragging=True, **kwargs):
# init PointBrowser
super(FindCenters,self).__init__(image,[[None,None]],**kwargs);
self.axis.set_title('FitHexagonCenters: %s' % self.imginfo['desc']);
self.fig.subplots_adjust(bottom=0.2); # space for sliders
# add slider for neighborhood size
self.nbhd_size = 5; # neighborhood size
axNbhd = self.fig.add_axes([0.2, 0.05, 0.1, 0.04]);
self.sNbhd = Slider(axNbhd,'neighbors ',2,20,valinit=self.nbhd_size,\
valfmt=' (%d)',dragging=dragging);
self.sNbhd.on_changed(self.ChangeNeighborhood);
# add slider for number of points
self.num_points = 1e6; # number of local maxima to find
axNum = self.fig.add_axes([0.45, 0.05, 0.3, 0.04]);
self.sNum = Slider(axNum, 'points ',0,100,valinit=self.num_points,\
valfmt=' (%d)',dragging=dragging);
self.sNum.on_changed(self.ChangeMaxPoints);
# add buttons
axRefine = self.fig.add_axes([0.85, 0.7, 0.1, 0.04]);
self.bRefine = Button(axRefine,'Refine');
self.bRefine.on_clicked(self.RefineCenters);
# initial calculation of local maximas
self.ChangeNeighborhood(self.nbhd_size);
def ChangeNeighborhood(self,val):
#print "ChangeNeighborhood"
self.nbhd_size = int(val);
# run initial peak fit
maxima,diff = self.find_local_maxima(self.image,self.nbhd_size);
# update max-number of points in points slider
self.sNum.valmax = Nmax = np.sum(maxima); # number of local max
self.sNum.ax.set_xlim((self.sNum.valmin, Nmax)); # rescale slider
self.sNum.set_val(min(self.num_points,Nmax)); # update value (calls ChangeMaxPoints)
def ChangeMaxPoints(self,val):
#print "ChangeMaxPoints()";
self.num_points = int(val);
self.points = self.refine_local_maxima(self.num_points);
self._update_points();
def RefineCenters(self,event):
" refine positions by fitting 2D Gaussian in neighborhood of local max "
from scipy.optimize import leastsq
from sys import stdout
#print "Refine()";
NN = self.nbhd_size;
Nx,Ny = self.image.shape;
dx,dy = np.mgrid[-NN:NN+1,-NN:NN+1];
# refine each point separately
self.points = self.points.astype(float); # allow subpixel precision
for ip in range(len(self.points)):
P = self.points[ip];
x,y = np.round(P);
# get neighborhood (skip border)
xmin,xmax = dx[[0,-1],0]+x; # first and last element in dx
ymin,ymax = dy[0,[0,-1]]+y; # " dy
if xmin<0 or ymin<0 or xmax>=Nx or ymax>=Ny: continue
nbhd = self.image[xmin:xmax+1,ymin:ymax+1];
assert nbhd.shape == (2*NN+1,2*NN+1)
# calculate center of mass
def gauss(x0,y0,A,B,fwhm):
return A*np.exp( - ((dx-x0)**2+(dy-y0)**2) / fwhm**2) + B;
p0 = (0.,0.,self.image[tuple(P)],0.,NN/2); # initial guess
residuals = lambda param: (nbhd - gauss(*param)).flat; # residuals
p,ierr = leastsq(lambda p: (nbhd - gauss(*p)).flat, p0);# least-squares fit
self.points[ip] = (x+p[0],y+p[1]); # correct position of point
# DEBUG: plot fits for each point
if self.verbosity > 0:
print "Refining Points... %d %%\r" % (100*ip/len(self.points-1)),
if self.verbosity > 3:
print "IN: ",p0
print "OUT: ",p
if self.verbosity > 10:
plt.figure();
ix = nbhd.shape[0]/2;
plt.plot(dy[ix],nbhd[ix], 'k',label='image');
plt.plot(dy[ix],gauss(*p0)[ix],'g',label='first guess');
plt.plot(dy[ix],gauss(*p)[ix], 'r',label='final fit');
plt.plot(dx[:,ix],nbhd[:,ix], 'k--');
plt.plot(dx[:,ix],gauss(*p0)[:,ix],'g--');
plt.plot(dx[:,ix],gauss(*p)[:,ix], 'r--');
plt.legend();
plt.show();
if self.verbosity > 0: print "Refining Points. Finished.";
stdout.flush();
self._update_points();
def find_local_maxima(self, data, neighborhood_size):
"""
find local maxima within neighborhood
idea from http://stackoverflow.com/questions/9111711
(get-coordinates-of-local-maxima-in-2d-array-above-certain-value)
"""
# find local maxima in image (width specified by neighborhood_size)
data_max = filters.maximum_filter(data,neighborhood_size);
maxima = (data == data_max);
assert np.sum(maxima) > 0; # we should always find local maxima
# remove connected pixels (plateaus)
labeled, num_objects = ndimage.label(maxima)
slices = ndimage.find_objects(labeled)
maxima *= 0;
for dx,dy in slices:
maxima[(dx.start+dx.stop-1)/2, (dy.start+dy.stop-1)/2] = 1
# calculate difference between local maxima and lowest
# pixel in neighborhood (will be used in select_local_maxima)
data_min = filters.minimum_filter(data,neighborhood_size);
diff = data_max - data_min;
self._maxima = maxima;
self._diff = diff;
return maxima,diff
def refine_local_maxima(self,N):
" select highest N local maxima using thresholding "
maxima = self._maxima; diff = self._diff;
# select highest local maxima using thresholding
if np.sum(maxima) > N:
# calc treshold from sorted list of differences for local maxima
thresh = np.sort(diff[maxima].flat)[-N];
# keep only maxima with diff>thresh
maxima = np.logical_and(maxima, diff>thresh);
# TODO: refine fit by local 2D Gauss-Fit
# return list of x,y positions of local maxima
return np.asarray(np.where(maxima)).T;
class BasicDendrogramViewer(object):
def __init__(self, dendrogram):
if dendrogram.data.ndim not in [2, 3]:
raise ValueError("Only 2- and 3-dimensional arrays are supported at this time")
self.array = dendrogram.data
self.dendrogram = dendrogram
self.plotter = DendrogramPlotter(dendrogram)
self.plotter.sort(reverse=True)
# Get the lines as individual elements, and the mapping from line to structure
self.lines = self.plotter.get_lines()
# Define the currently selected subtree
self.selected = None
self.selected_lines = None
self.selected_contour = None
# Initiate plot
import matplotlib.pyplot as plt
self.fig = plt.figure(figsize=(14, 8))
self.ax1 = self.fig.add_axes([0.1, 0.1, 0.4, 0.7])
from matplotlib.widgets import Slider
self._clim = (np.min(self.array[~np.isnan(self.array) & ~np.isinf(self.array)]),
np.max(self.array[~np.isnan(self.array) & ~np.isinf(self.array)]))
if self.array.ndim == 2:
self.slice = None
self.image = self.ax1.imshow(self.array, origin='lower', interpolation='nearest', vmin=self._clim[0], vmax=self._clim[1], cmap=plt.cm.gray)
else:
self.slice = int(round(self.array.shape[0] / 2.))
self.image = self.ax1.imshow(self.array[self.slice, :, :], origin='lower', interpolation='nearest', vmin=self._clim[0], vmax=self._clim[1], cmap=plt.cm.gray)
self.slice_slider_ax = self.fig.add_axes([0.1, 0.95, 0.4, 0.03])
self.slice_slider_ax.set_xticklabels("")
self.slice_slider_ax.set_yticklabels("")
self.slice_slider = Slider(self.slice_slider_ax, "3-d slice", 0, self.array.shape[0], valinit=self.slice, valfmt="%i")
self.slice_slider.on_changed(self.update_slice)
self.slice_slider.drawon = False
self.vmin_slider_ax = self.fig.add_axes([0.1, 0.90, 0.4, 0.03])
self.vmin_slider_ax.set_xticklabels("")
self.vmin_slider_ax.set_yticklabels("")
self.vmin_slider = Slider(self.vmin_slider_ax, "vmin", self._clim[0], self._clim[1], valinit=self._clim[0])
self.vmin_slider.on_changed(self.update_vmin)
self.vmin_slider.drawon = False
self.vmax_slider_ax = self.fig.add_axes([0.1, 0.85, 0.4, 0.03])
self.vmax_slider_ax.set_xticklabels("")
self.vmax_slider_ax.set_yticklabels("")
self.vmax_slider = Slider(self.vmax_slider_ax, "vmax", self._clim[0], self._clim[1], valinit=self._clim[1])
self.vmax_slider.on_changed(self.update_vmax)
self.vmax_slider.drawon = False
self.ax2 = self.fig.add_axes([0.6, 0.3, 0.35, 0.4])
self.ax2.add_collection(self.lines)
self.selected_label = self.fig.text(0.6, 0.75, "No structure selected", fontsize=18)
x = [p.vertices[:, 0] for p in self.lines.get_paths()]
y = [p.vertices[:, 1] for p in self.lines.get_paths()]
xmin = np.min(x)
xmax = np.max(x)
ymin = np.min(y)
ymax = np.max(y)
self.lines.set_picker(2.)
dx = xmax - xmin
self.ax2.set_xlim(xmin - dx * 0.1, xmax + dx * 0.1)
self.ax2.set_ylim(ymin * 0.5, ymax * 2.0)
self.ax2.set_yscale('log')
self.fig.canvas.mpl_connect('pick_event', self.line_picker)
self.fig.canvas.mpl_connect('button_press_event', self.select_from_map)
plt.show()
def update_slice(self, pos=None):
if self.array.ndim == 2:
self.image.set_array(self.array)
else:
self.slice = int(round(pos))
self.image.set_array(self.array[self.slice,:,:])
self.remove_contour()
self.update_contour()
self.fig.canvas.draw()
def update_vmin(self, vmin):
if vmin > self._clim[1]:
self._clim = (self._clim[1], self._clim[1])
else:
self._clim = (vmin, self._clim[1])
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def update_vmax(self, vmax):
if vmax < self._clim[0]:
self._clim = (self._clim[0], self._clim[0])
else:
self._clim = (self._clim[0], vmax)
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def select_from_map(self, event):
# Only do this if no tools are currently selected
if event.canvas.toolbar.mode != '':
return
if event.inaxes is self.ax1:
# Find pixel co-ordinates of click
ix = int(round(event.xdata))
iy = int(round(event.ydata))
if self.array.ndim == 2:
indices = (iy, ix)
else:
indices = (self.slice, iy, ix)
# Select the structure
structure = self.dendrogram.node_at(indices)
self.select(structure)
# Re-draw
event.canvas.draw()
def line_picker(self, event):
# Only do this if no tools are currently selected
if event.canvas.toolbar.mode != '':
return
# event.ind gives the indices of the paths that have been selected
# Find levels of selected paths
peaks = [event.artist.structures[i].get_peak(subtree=True)[1] for i in event.ind]
# Find position of minimum level (may be duplicates, let Numpy decide)
ind = event.ind[np.argmax(peaks)]
# Extract structure
structure = event.artist.structures[ind]
# If 3-d, select the slice
if self.array.ndim == 3:
peak_index = structure.get_peak(subtree=True)
self.slice_slider.set_val(peak_index[0][0])
# Select the structure
self.select(structure)
# Re-draw
event.canvas.draw()
def select(self, structure):
# Remove previously selected collection
if self.selected_lines is not None:
self.ax2.collections.remove(self.selected_lines)
self.selected_lines = None
self.remove_contour()
if structure is None:
self.selected_label.set_text("No structure selected")
self.fig.canvas.draw()
return
self.selected = structure
self.selected_label.set_text("Selected structure: {0}".format(structure.idx))
# Get collection for this substructure
self.selected_lines = self.plotter.get_lines(structure=structure)
self.selected_lines.set_color('red')
self.selected_lines.set_linewidth(2)
self.selected_lines.set_alpha(0.5)
# Add to axes
self.ax2.add_collection(self.selected_lines)
self.update_contour()
def remove_contour(self):
if self.selected_contour is not None:
for collection in self.selected_contour.collections:
self.ax1.collections.remove(collection)
self.selected_contour = None
def update_contour(self):
if self.selected is not None:
mask = self.selected.get_mask(self.array.shape, subtree=True)
if self.array.ndim == 3:
mask = mask[self.slice, :, :]
self.selected_contour = self.ax1.contour(mask, colors='red', linewidths=2, levels=[0.5], alpha=0.5)
class viscm_editor(object):
def __init__(self, min_Jp=15, max_Jp=95, xp=None, yp=None):
from .bezierbuilder import BezierModel, BezierBuilder
axes = _viscm_editor_axes()
ax_btn_wireframe = plt.axes([0.7, 0.15, 0.1, 0.025])
self.btn_wireframe = Button(ax_btn_wireframe, 'Show 3D gamut')
self.btn_wireframe.on_clicked(self.plot_3d_gamut)
ax_btn_wireframe = plt.axes([0.81, 0.15, 0.1, 0.025])
self.btn_save = Button(ax_btn_wireframe, 'Save colormap')
self.btn_save.on_clicked(self.save_colormap)
ax_btn_props = plt.axes([0.81, 0.1, 0.1, 0.025])
self.btn_props = Button(ax_btn_props, 'Properties')
self.btn_props.on_clicked(self.show_viscm)
self.prop_windows = []
axcolor = 'None'
ax_jp_min = plt.axes([0.1, 0.1, 0.5, 0.03], axisbg=axcolor)
ax_jp_min.imshow(np.linspace(0, 100, 101).reshape(1, -1), cmap='gray')
ax_jp_min.set_xlim(0, 100)
ax_jp_max = plt.axes([0.1, 0.15, 0.5, 0.03], axisbg=axcolor)
ax_jp_max.imshow(np.linspace(0, 100, 101).reshape(1, -1), cmap='gray')
self.jp_min_slider = Slider(ax_jp_min, r"$J'_\mathrm{min}$", 0, 100, valinit=min_Jp)
self.jp_max_slider = Slider(ax_jp_max, r"$J'_\mathrm{max}$", 0, 100, valinit=max_Jp)
self.jp_min_slider.on_changed(self._jp_update)
self.jp_max_slider.on_changed(self._jp_update)
# This is my favorite set of control points so far (just from playing
# around with things):
# min_Jp = 15
# max_Jp = 95
# xp =
# [-4, 27.041103603603631, 84.311067635550557, 12.567076579094476, -9.6]
# yp =
# [-34, -41.447876447876524, 36.28563443264386, 25.357741755170423, 41]
# -- njs, 2015-04-05
if xp is None:
xp = [-4, 38.289146128951984, 52.1923711457504,
39.050944362271053, 18.60872492130315, -9.6]
if yp is None:
yp = [-34, -34.34528254916614, -21.594701710471412,
31.701084689194829, 29.510846891948262, 41]
self.bezier_model = BezierModel(xp, yp)
self.cmap_model = BezierCMapModel(self.bezier_model,
self.jp_min_slider.val,
self.jp_max_slider.val)
self.highlight_point_model = HighlightPointModel(self.cmap_model, 0.5)
self.bezier_builder = BezierBuilder(axes['bezier'], self.bezier_model)
self.bezier_gamut_viewer = GamutViewer2D(axes['bezier'],
self.highlight_point_model)
tmp = HighlightPoint2DView(axes['bezier'],
self.highlight_point_model)
self.bezier_highlight_point_view = tmp
#draw_pure_hue_angles(axes['bezier'])
axes['bezier'].set_xlim(-100, 100)
axes['bezier'].set_ylim(-100, 100)
self.cmap_view = CMapView(axes['cm'], self.cmap_model)
self.cmap_highlighter = HighlightPointBuilder(
axes['cm'],
self.highlight_point_model)
print("Click sliders at bottom to change min/max lightness")
print("Click on colorbar to adjust gamut view")
print("Click-drag to move control points, ")
print(" shift-click to add, control-click to delete")
def plot_3d_gamut(self, event):
fig, ax = plt.subplots(subplot_kw=dict(projection='3d'))
self.wireframe_view = WireframeView(ax,
self.cmap_model,
self.highlight_point_model)
plt.show()
def save_colormap(self, event):
import textwrap
template = textwrap.dedent('''
from matplotlib.colors import LinearSegmentedColormap
from numpy import nan, inf
# Used to reconstruct the colormap in pycam02ucs.cm.viscm
parameters = {{'xp': {xp},
'yp': {yp},
'min_Jp': {min_Jp},
'max_Jp': {max_Jp}}}
cm_data = {array_list}
test_cm = LinearSegmentedColormap.from_list(__file__, cm_data)
if __name__ == "__main__":
import matplotlib.pyplot as plt
import numpy as np
try:
from pycam02ucs.cm.viscm import viscm
viscm(test_cm)
except ImportError:
print("pycam02ucs not found, falling back on simple display")
plt.imshow(np.linspace(0, 100, 256)[None, :], aspect='auto',
cmap=test_cm)
plt.show()
''')
rgb, _ = self.cmap_model.get_sRGB(num=256)
with open('/tmp/new_cm.py', 'w') as f:
array_list = np.array_repr(rgb, max_line_width=78)
array_list = array_list.replace('array(', '')[:-1]
xp, yp = self.cmap_model.bezier_model.get_control_points()
data = dict(array_list=array_list,
xp=xp,
yp=yp,
min_Jp=self.cmap_model.min_Jp,
max_Jp=self.cmap_model.max_Jp)
f.write(template.format(**data))
print("*" * 50)
print("Saved colormap to /tmp/new_cm.py")
print("*" * 50)
def show_viscm(self, event):
cm = LinearSegmentedColormap.from_list(
'test_cm',
self.cmap_model.get_sRGB(num=256)[0])
self.prop_windows.append(viscm(cm, name='test_cm'))
plt.show()
def _jp_update(self, val):
jp_min = self.jp_min_slider.val
jp_max = self.jp_max_slider.val
smallest, largest = min(jp_min, jp_max), max(jp_min, jp_max)
if (jp_min > smallest) or (jp_max < largest):
self.jp_min_slider.set_val(smallest)
self.jp_max_slider.set_val(largest)
self.cmap_model.set_Jp_minmax(smallest, largest)
class GUI(animation.TimedAnimation):
"""
interface for viewing a movie and its associated roi, traces, other things
implementation is only through matplotlib, and is non-blocking
backend affects performance somewhat dramatically. have achieved decent performance with qt5agg and tkagg
"""
def __init__(self, mov, roi, traces, images={}, cmap=pl.cm.viridis, **kwargs):
"""
Parameters:
mov : 3d np array, 0'th axis is time/frames
roi : 3d np array, one roi per item in 0'th axis, each of which is a True/False mask indicating roi (True=inside roi)
traces : 2d np array, 0'th axis is time, 1st axis is sources
images: dictionary of still images
Attributes:
roi_kept : boolean array of length of supplied roi, indicating whether or not roi should be kept based on user input
"""
self.mov = mov
self.roi_idxs = np.array([np.argwhere(r.flat).squeeze() for r in roi])
self.roi_centers = np.array([np.mean(np.argwhere(r),axis=0) for r in roi])
self.roi_orig = roi.copy()
self.roi = pretty_roi(roi)
self.roi_kept = np.ones(len(self.roi_idxs)).astype(bool)
self.traces = traces
self.images = images
# figure setup
self.cmap = cmap
self.fig = pl.figure()
NR,NC = 128,32
gs = gridspec.GridSpec(nrows=NR, ncols=NC)
gs.update(wspace=0.1, hspace=0.1, left=.04, right=.96, top=.98, bottom=.02)
# movie axes
self.ax_contrast0 = self.fig.add_subplot(gs[0:5,0:NC//3])
self.ax_contrast1 = self.fig.add_subplot(gs[5:10,0:NC//3])
self.ax_mov = self.fig.add_subplot(gs[10:55,0:NC//3])
self.ax_mov.axis('off')
self.ax_img = self.fig.add_subplot(gs[55:100,0:NC//3])
self.ax_img.axis('off')
self.axs_imbuts = [self.fig.add_subplot(gs[110:128,idx*2:idx*2+2]) for idx,i in enumerate(self.images)]
# trace axes
self.ax_trcs = self.fig.add_subplot(gs[0:64,NC//2:])
self.ax_trc = self.fig.add_subplot(gs[65:85,NC//2:])
self.ax_trc.set_xlim([0, len(self.traces)])
self.ax_nav = self.fig.add_subplot(gs[85:90,NC//2:])
self.ax_nav.set_xlim([0, len(self.traces)])
self.ax_nav.axis('off')
self.ax_rm = self.fig.add_subplot(gs[95:110,NC//2:])
# interactivity
self.c0,self.c1= 0,100
self.sl_contrast0 = Slider(self.ax_contrast0, 'Low', 0., 255.0, valinit=self.c0, valfmt='%d')
self.sl_contrast1 = Slider(self.ax_contrast1, 'Hi', 0., 255.0, valinit=self.c1, valfmt='%d')
self.sl_contrast0.on_changed(self.evt_contrast)
self.sl_contrast1.on_changed(self.evt_contrast)
self.img_buttons = [Button(ax,k) for k,ax in zip(list(self.images.keys()),self.axs_imbuts)]
self.but_rm = Button(self.ax_rm, 'Remove All ROIs Currently in FOV')
# display initial things
self.movdata = self.ax_mov.imshow(self.mov[0])
self.movdata.set_animated(True)
self.roidata = self.ax_mov.imshow(self.roi, cmap=self.cmap, alpha=0.5, vmin=np.nanmin(self.roi), vmax=np.nanmax(self.roi))
self.trdata, = self.ax_trc.plot(np.zeros(len(self.traces)))
self.navdata, = self.ax_nav.plot([-2,-2],[-1,np.max(self.traces)], 'r-')
if len(self.images):
lab,im = list(self.images.items())[0]
self.imgdata = self.ax_img.imshow(im)
self.ax_img.set_ylabel(lab)
self.plot_current_traces()
# callbacks
for ib,lab in zip(self.img_buttons,list(self.images.keys())):
ib.on_clicked(lambda evt, lab=lab: self.evt_imbut(evt,lab))
self.but_rm.on_clicked(self.remove_roi)
self.fig.canvas.mpl_connect('button_press_event', self.evt_click)
self.ax_mov.callbacks.connect('xlim_changed', self.evt_zoom)
self.ax_mov.callbacks.connect('ylim_changed', self.evt_zoom)
# runtime
self._idx = -1
self.t0 = time.clock()
self.always_draw = [self.movdata, self.roidata, self.navdata]
self.blit_clear_axes = [self.ax_mov, self.ax_nav]
# parent init
animation.TimedAnimation.__init__(self, self.fig, interval=40, blit=True, **kwargs)
@property
def frame_seq(self):
#print (time.clock()-self.t0)
self._idx += 1
if self._idx == len(self.mov):
self._idx = 0
self.navdata.set_xdata([self._idx, self._idx])
yield self.mov[self._idx]
@frame_seq.setter
def frame_seq(self, val):
pass
def new_frame_seq(self):
return self.mov
def _init_draw(self):
self._draw_frame(self.mov[0])
self._drawn_artists = self.always_draw
def _draw_frame(self, d):
self.t0 = time.clock()
self.movdata.set_data(d)
# blit
self._drawn_artists = self.always_draw
for da in self._drawn_artists:
da.set_animated(True)
def _blit_clear(self, artists, bg_cache):
for ax in self.blit_clear_axes:
if ax in bg_cache:
self.fig.canvas.restore_region(bg_cache[ax])
def evt_contrast(self, val):
self.c0 = self.sl_contrast0.val
self.c1 = self.sl_contrast1.val
if self.c0 > self.c1:
self.c0 = self.c1-1
self.sl_contrast0.set_val(self.c0)
if self.c1 < self.c0:
self.c1 = self.c0+1
self.sl_contrast1.set_val(self.c1)
self.movdata.set_clim(vmin=self.c0, vmax=self.c1)
self.imgdata.set_clim(vmin=self.c0, vmax=self.c1)
def evt_imbut(self, evt, lab):
self.imgdata.set_data(self.images[lab])
self.ax_img.set_title(lab)
def evt_click(self, evt):
if not evt.inaxes:
return
elif evt.inaxes == self.ax_mov:
# select roi
x,y = int(np.round(evt.xdata)), int(np.round(evt.ydata))
idx = np.ravel_multi_index((y,x), self.roi.shape)
inside = np.argwhere([idx in ri for ri in self.roi_idxs])
if len(inside)==0:
return
i = inside[0]
self.set_current_trace(i)
elif evt.inaxes in [self.ax_nav]:
x = int(np.round(evt.xdata))
self._idx = x
def evt_zoom(self, *args):
self.plot_current_traces()
def set_current_trace(self, idx):
col = self.cmap(np.linspace(0,1,np.sum(self.roi_kept)))[np.squeeze(idx)]
t = self.traces[:,idx]
self.trdata.set_ydata(t)
self.trdata.set_color(col)
self.ax_trc.set_ylim([t.min(), t.max()])
self.ax_trc.set_title('ROI {}'.format(idx))
self.ax_trc.figure.canvas.draw()
def get_current_roi(self):
croi = np.array([isin(rc,self.ax_mov) for rc in self.roi_centers])
croi[self.roi_kept==False] = False
return croi
def remove_roi(self, evt):
self.current_roi = self.get_current_roi()
self.roi_kept[self.current_roi] = False
# update
if np.sum(self.roi_kept):
proi = pretty_roi(self.roi_orig[self.roi_kept])
self.roidata.set_data(proi)
self.roidata.set_clim(vmin=np.nanmin(proi), vmax=np.nanmax(proi))
else:
self.roidata.remove()
def plot_current_traces(self):
self.current_roi = self.get_current_roi()
if np.sum(self.current_roi)==0:
return
for line in self.ax_trcs.get_lines():
line.remove()
cols = self.cmap(np.linspace(0,1,len(self.roi_idxs)))[self.current_roi]
lastmax = 0
for t,c in zip(self.traces.T[self.current_roi],cols):
self.ax_trcs.plot((t-t.min())+lastmax, color=c)
lastmax = np.max(t)
self.ax_trcs.set_ylim([0,lastmax])
def view_patches_bar(Yr, A, C, b, f, d1, d2, YrA=None, img=None):
"""view spatial and temporal components interactively
Parameters:
-----------
Yr: np.ndarray
movie in format pixels (d) x frames (T)
A: sparse matrix
matrix of spatial components (d x K)
C: np.ndarray
matrix of temporal components (K x T)
b: np.ndarray
spatial background (vector of length d)
f: np.ndarray
temporal background (vector of length T)
d1,d2: np.ndarray
frame dimensions
YrA: np.ndarray
ROI filtered residual as it is given from update_temporal_components
If not given, then it is computed (K x T)
img: np.ndarray
background image for contour plotting. Default is the image of all spatial components (d1 x d2)
"""
pl.ion()
if 'csc_matrix' not in str(type(A)):
A = csc_matrix(A)
if 'array' not in str(type(b)):
b = b.toarray()
nr, T = C.shape
nb = f.shape[0]
nA2 = np.sqrt(np.array(A.power(2).sum(axis=0))).squeeze()
if YrA is None:
Y_r = spdiags(old_div(1, nA2), 0, nr, nr) * (A.T.dot(Yr) -
(A.T.dot(b)).dot(f) - (A.T.dot(A)).dot(C)) + C
else:
Y_r = YrA + C
if img is None:
img = np.reshape(np.array(A.mean(axis=1)), (d1, d2), order='F')
fig = pl.figure(figsize=(10, 10))
axcomp = pl.axes([0.05, 0.05, 0.9, 0.03])
ax1 = pl.axes([0.05, 0.55, 0.4, 0.4])
ax3 = pl.axes([0.55, 0.55, 0.4, 0.4])
ax2 = pl.axes([0.05, 0.1, 0.9, 0.4])
s_comp = Slider(axcomp, 'Component', 0, nr + nb - 1, valinit=0)
vmax = np.percentile(img, 95)
def update(val):
i = np.int(np.round(s_comp.val))
print(('Component:' + str(i)))
if i < nr:
ax1.cla()
imgtmp = np.reshape(A[:, i].toarray(), (d1, d2), order='F')
ax1.imshow(imgtmp, interpolation='None', cmap=pl.cm.gray, vmax=np.max(imgtmp)*0.5)
ax1.set_title('Spatial component ' + str(i + 1))
ax1.axis('off')
ax2.cla()
ax2.plot(np.arange(T), Y_r[i], 'c', linewidth=3)
ax2.plot(np.arange(T), C[i], 'r', linewidth=2)
ax2.set_title('Temporal component ' + str(i + 1))
ax2.legend(labels=['Filtered raw data', 'Inferred trace'])
ax3.cla()
ax3.imshow(img, interpolation='None', cmap=pl.cm.gray, vmax=vmax)
imgtmp2 = imgtmp.copy()
imgtmp2[imgtmp2 == 0] = np.nan
ax3.imshow(imgtmp2, interpolation='None',
alpha=0.5, cmap=pl.cm.hot)
ax3.axis('off')
else:
ax1.cla()
bkgrnd = np.reshape(b[:, i - nr], (d1, d2), order='F')
ax1.imshow(bkgrnd, interpolation='None')
ax1.set_title('Spatial background ' + str(i + 1 - nr))
ax1.axis('off')
ax2.cla()
ax2.plot(np.arange(T), np.squeeze(np.array(f[i - nr, :])))
ax2.set_title('Temporal background ' + str(i + 1 - nr))
def arrow_key_image_control(event):
if event.key == 'left':
new_val = np.round(s_comp.val - 1)
if new_val < 0:
new_val = 0
s_comp.set_val(new_val)
elif event.key == 'right':
new_val = np.round(s_comp.val + 1)
if new_val > nr + nb:
new_val = nr + nb
s_comp.set_val(new_val)
else:
pass
s_comp.on_changed(update)
s_comp.set_val(0)
fig.canvas.mpl_connect('key_release_event', arrow_key_image_control)
pl.show()
RmaxV = Slider(Rmax, 'Rmax', 1, 254, valinit=rgbinit[0])
RminV = Slider(Rmin, 'Rmin', 1, 254, valinit=rgbinit[1])
GmaxV = Slider(Gmax, 'Gmax', 1, 254, valinit=rgbinit[2])
GminV = Slider(Gmin, 'Gmin', 1, 254, valinit=rgbinit[3])
BmaxV = Slider(Bmax, 'Bmax', 1, 254, valinit=rgbinit[4])
BminV = Slider(Bmin, 'Bmin', 1, 254, valinit=rgbinit[5])
RmaxV.on_changed(sliceupdateRmax)
RminV.on_changed(sliceupdateRmin)
GmaxV.on_changed(sliceupdateGmax)
GminV.on_changed(sliceupdateGmin)
BmaxV.on_changed(sliceupdateBmax)
BminV.on_changed(sliceupdateBmin)
ff=1
else:
RmaxV.set_val(rgbinit[0])
RminV.set_val(rgbinit[1])
GmaxV.set_val(rgbinit[2])
GminV.set_val(rgbinit[3])
BmaxV.set_val(rgbinit[4])
BminV.set_val(rgbinit[5])
file = open("rgb.txt", "w")
file.write(str(rgbinit))
file.close()
print rgbinit
plt.pause(0.001)
plt.show(block=False)
#print samp.val,sfreq.val
def view_patches_bar(Yr, A, C, b, f, d1, d2, YrA=None, secs=1, img=None):
"""view spatial and temporal components interactively
Parameters
-----------
Yr: np.ndarray
movie in format pixels (d) x frames (T)
A: sparse matrix
matrix of spatial components (d x K)
C: np.ndarray
matrix of temporal components (K x T)
b: np.ndarray
spatial background (vector of length d)
f: np.ndarray
temporal background (vector of length T)
d1,d2: np.ndarray
frame dimensions
YrA: np.ndarray
ROI filtered residual as it is given from update_temporal_components
If not given, then it is computed (K x T)
img: np.ndarray
background image for contour plotting. Default is the image of all spatial components (d1 x d2)
"""
plt.ion()
nr, T = C.shape
A2 = A.copy()
A2.data **= 2
nA2 = np.sqrt(np.array(A2.sum(axis=0))).squeeze()
#A = A*spdiags(1/nA2,0,nr,nr)
#C = spdiags(nA2,0,nr,nr)*C
b = np.squeeze(b)
f = np.squeeze(f)
if YrA is None:
Y_r = np.array(A.T * np.matrix(Yr) - (A.T * np.matrix(b[:, np.newaxis])) * np.matrix(
f[np.newaxis]) - (A.T.dot(A)) * np.matrix(C) + C)
else:
Y_r = YrA + C
A = A * spdiags(1 / nA2, 0, nr, nr)
A = A.todense()
imgs = np.reshape(np.array(A), (d1, d2, nr), order='F')
if img is None:
img = np.mean(imgs[:, :, :-1], axis=-1)
bkgrnd = np.reshape(b, (d1, d2), order='F')
fig = plt.figure(figsize=(10, 10))
axcomp = plt.axes([0.05, 0.05, 0.9, 0.03])
ax1 = plt.axes([0.05, 0.55, 0.4, 0.4])
# ax1.axis('off')
ax3 = plt.axes([0.55, 0.55, 0.4, 0.4])
# ax1.axis('off')
ax2 = plt.axes([0.05, 0.1, 0.9, 0.4])
# axcolor = 'lightgoldenrodyellow'
# axcomp = plt.axes([0.25, 0.1, 0.65, 0.03], axisbg=axcolor)
s_comp = Slider(axcomp, 'Component', 0, nr, valinit=0)
vmax = np.percentile(img, 98)
def update(val):
i = np.int(np.round(s_comp.val))
print 'Component:' + str(i)
if i < nr:
ax1.cla()
imgtmp = imgs[:, :, i]
ax1.imshow(imgtmp, interpolation='None', cmap=plt.cm.gray)
ax1.set_title('Spatial component ' + str(i + 1))
ax1.axis('off')
ax2.cla()
ax2.plot(np.arange(T), np.squeeze(np.array(Y_r[i, :])), 'c', linewidth=3)
ax2.plot(np.arange(T), np.squeeze(np.array(C[i, :])), 'r', linewidth=2)
ax2.set_title('Temporal component ' + str(i + 1))
ax2.legend(labels=['Filtered raw data', 'Inferred trace'])
ax3.cla()
ax3.imshow(img, interpolation='None', cmap=plt.cm.gray, vmax=vmax)
imgtmp2 = imgtmp.copy()
imgtmp2[imgtmp2 == 0] = np.nan
ax3.imshow(imgtmp2, interpolation='None', alpha=0.5, cmap=plt.cm.hot)
else:
ax1.cla()
ax1.imshow(bkgrnd, interpolation='None')
ax1.set_title('Spatial background background')
ax2.cla()
ax2.plot(np.arange(T), np.squeeze(np.array(f)))
ax2.set_title('Temporal background')
def arrow_key_image_control(event):
if event.key == 'left':
new_val = np.round(s_comp.val - 1)
if new_val < 0:
new_val = 0
s_comp.set_val(new_val)
elif event.key == 'right':
new_val = np.round(s_comp.val + 1)
if new_val > nr:
new_val = nr
s_comp.set_val(new_val)
else:
pass
s_comp.on_changed(update)
s_comp.set_val(0)
id2 = fig.canvas.mpl_connect('key_release_event', arrow_key_image_control)
plt.show()
class GridCircle:
"""
Provided a 2D grid, this object:
> plots the grid with a circle which centre and radius are adjustable,
> plots values of the grid along the circle alongside the grid.
"""
def __init__(self,
grid,
extent=(-1, 1, -1, 1),
circle_centre=(0, 0),
min=None,
max=None,
points_theta=100,
linear_interpolation=False,
show_slider=True):
"""
Parameters
----------
grid : 2D array-like
Grid to plot values from.
extent : scalars (left, right, bottom, top)
Values of space variables at corners. (default: (-1, 1, -1, 1))
circle_centre : scalars (x, y)
Location of the centre of the circle to draw on top of grid.
min : float
Minimum value for the colormap. (default: None)
NOTE: None will be considered as the minimum being the minimum
value of grid.
max : float
Maximum value for the colormap. (default: None)
NOTE: None will be considered as the maximum being the maximum
value of grid.
points_theta : int
Number of points to consider in the interval [0, 2\\pi] when
computing values along circle.
linear_interpolation : bool
Get value by linear interpolation of neighbouring grid boxes.
(default: False)
show_slider : bool
Display circle radius slider.
"""
self.circle_centre = np.array(circle_centre)
self.radius = 0 # radius of the circle
self.points_theta = points_theta
self.theta = np.linspace(0, 2 * np.pi, points_theta)
self.linear_interpolation = linear_interpolation
self.show_slider = show_slider
self.fig, (self.ax_grid, self.ax_plot) = plt.subplots(
1, 2
) # matplotlib.figure.Figure object and matplotlib.axes.Axes objects for grid and value plot
# COLORMAP
self.min = np.min(grid) if min == None else min
self.max = np.max(grid) if max == None else max
self.norm = colors.Normalize(vmin=self.min,
vmax=self.max) # normalises data
self.scalarmap = cmx.ScalarMappable(
norm=self.norm, cmap=cmap) # scalar map for grid values
# PLOT
self.ax_plot.set_ylim(
[self.min,
self.max]) # setting y-axis limit of plot as grid extrema
self.ax_plot.set_xlim([0, 2 * np.pi]) # angle on the cirlce
self.line, = self.ax_plot.plot(
np.linspace(0, 2 * np.pi, self.points_theta),
[0] * self.points_theta) # plot of values along circle
# SLIDER
self.extent = np.array(extent) # grid extent
if self.show_slider:
self.slider_ax = make_axes_locatable(self.ax_plot).append_axes(
'bottom', size='5%', pad=0.5) # slider axes
self.slider = Slider(self.slider_ax,
'radius',
0,
np.min(np.abs(self.extent)),
valinit=self.radius) # slider
self.slider.on_changed(
self.update_slider
) # call self.update_slider() on slider update
# GRID
#grid = np.array(grid)
self.grid_plot = self.ax_grid.imshow(grid,
cmap=cmap,
norm=self.norm,
extent=self.extent) # grid plot
self.colormap_ax = make_axes_locatable(self.ax_grid).append_axes(
'right', size='5%', pad=0.05) # color map axes
self.colormap = mpl.colorbar.ColorbarBase(
self.colormap_ax,
cmap=cmap,
norm=self.norm,
orientation='vertical') # color map
self.circle = plt.Circle(self.circle_centre,
self.radius,
color='black',
fill=False) # circle on grid
self.ax_grid.add_artist(self.circle)
self.ax_grid.figure.canvas.mpl_connect(
'button_press_event',
self.update_grid) # call self.update_grid() on button press event
self.update_grid_plot(grid) # plots grid and updates circle and plot
def get_fig_ax_cmap(self):
"""
Returns
-------
fig : matplotlib.pyplot.figure object
Figure.
(ax_grid, ax_plot) : matplotlib.axes.Axes tuple
Grid and plot axes.
colormap : matplotlib.colorbar.ColorbarBase object
Color map.
"""
return self.fig, (self.ax_grid, self.ax_plot), self.colormap
def update_grid_plot(self, grid, extent=None):
"""
Plots grid.
Parameters
----------
grid : 2D array-like
Grid to plot values from.
extent : scalars (left, right, bottom, top)
Values of space variables at corners. (default: None)
NOTE: None will be considered as extent to be self.extent.
"""
if extent != None: self.extent = np.array(extent)
self.grid = Grid(grid, extent=self.extent)
self.grid_plot.set_data(self.grid.grid) # plots grid
self.grid_plot.set_extent(self.extent) # set extent
self.draw() # updates circle and plot
def update_grid(self, event):
"""
Executes on click on figure.
Updates radius of cirlce on figure.
"""
if event.inaxes != self.circle.axes:
return # if Axes instance mouse is over is different than circle's figure Axes
self.radius = np.sqrt(
np.sum(
(np.array([event.xdata, event.ydata]) - self.circle_centre)**2)
) # radius set to distance between centre of circle and clicked point
self.slider.set_val(self.radius) # updates slider value
self.draw() # updates figure
def update_slider(self, event):
"""
Executes on slider change.
Updates radius of circle on figure.
"""
self.radius = self.slider.val # radius set to slider value
self.draw() # updates figure
def draw(self):
"""
Updates figure.
"""
self.line.set_ydata(
list(
map(
lambda angle: self.grid.get_value_polar(
self.radius,
angle,
centre=self.circle_centre,
linear_interpolation=self.linear_interpolation),
self.theta))) # values of the grid along the circle
self.circle.set_radius(self.radius) # adjusting circle radius
self.ax_grid.figure.canvas.draw() # updating grid
self.ax_plot.figure.canvas.draw() # updating plot
class BasicDendrogramViewer(object):
def __init__(self, dendrogram):
if dendrogram.data.ndim not in [2, 3]:
raise ValueError(
"Only 2- and 3-dimensional arrays are supported at this time")
self.hub = SelectionHub()
self._connect_to_hub()
self.array = dendrogram.data
self.dendrogram = dendrogram
self.plotter = DendrogramPlotter(dendrogram)
self.plotter.sort(reverse=True)
# Get the lines as individual elements, and the mapping from line to structure
self.lines = self.plotter.get_lines(edgecolor='k')
# Define the currently selected subtree
self.selected_lines = {}
self.selected_contour = {}
# The keys in these dictionaries are event button IDs.
# Initiate plot
import matplotlib.pyplot as plt
self.fig = plt.figure(figsize=(14, 8))
ax_image_limits = [0.1, 0.1, 0.4, 0.7]
try:
from wcsaxes import WCSAxes
__wcaxes_imported = True
except ImportError:
__wcaxes_imported = False
if self.dendrogram.wcs is not None:
warnings.warn("`WCSAxes` package required for wcs coordinate display.")
if self.dendrogram.wcs is not None and __wcaxes_imported:
if self.array.ndim == 2:
slices = ('x', 'y')
else:
slices = ('x', 'y', 1)
ax_image = WCSAxes(self.fig, ax_image_limits, wcs=self.dendrogram.wcs, slices=slices)
self.ax_image = self.fig.add_axes(ax_image)
else:
self.ax_image = self.fig.add_axes(ax_image_limits)
from matplotlib.widgets import Slider
self._clim = (np.min(self.array[~np.isnan(self.array) & ~np.isinf(self.array)]),
np.max(self.array[~np.isnan(self.array) & ~np.isinf(self.array)]))
if self.array.ndim == 2:
self.slice = None
self.image = self.ax_image.imshow(self.array, origin='lower', interpolation='nearest', vmin=self._clim[0], vmax=self._clim[1], cmap=plt.cm.gray)
self.slice_slider = None
else:
if self.array.shape[0] > 1:
self.slice = int(round(self.array.shape[0] / 2.))
self.slice_slider_ax = self.fig.add_axes([0.1, 0.95, 0.4, 0.03])
self.slice_slider_ax.set_xticklabels("")
self.slice_slider_ax.set_yticklabels("")
self.slice_slider = Slider(self.slice_slider_ax, "3-d slice", 0, self.array.shape[0], valinit=self.slice, valfmt="%i")
self.slice_slider.on_changed(self.update_slice)
self.slice_slider.drawon = False
else:
self.slice = 0
self.slice_slider = None
self.image = self.ax_image.imshow(self.array[self.slice, :,:], origin='lower', interpolation='nearest', vmin=self._clim[0], vmax=self._clim[1], cmap=plt.cm.gray)
self.vmin_slider_ax = self.fig.add_axes([0.1, 0.90, 0.4, 0.03])
self.vmin_slider_ax.set_xticklabels("")
self.vmin_slider_ax.set_yticklabels("")
self.vmin_slider = Slider(self.vmin_slider_ax, "vmin", self._clim[0], self._clim[1], valinit=self._clim[0])
self.vmin_slider.on_changed(self.update_vmin)
self.vmin_slider.drawon = False
self.vmax_slider_ax = self.fig.add_axes([0.1, 0.85, 0.4, 0.03])
self.vmax_slider_ax.set_xticklabels("")
self.vmax_slider_ax.set_yticklabels("")
self.vmax_slider = Slider(self.vmax_slider_ax, "vmax", self._clim[0], self._clim[1], valinit=self._clim[1])
self.vmax_slider.on_changed(self.update_vmax)
self.vmax_slider.drawon = False
self.ax_dendrogram = self.fig.add_axes([0.6, 0.3, 0.35, 0.4])
self.ax_dendrogram.add_collection(self.lines)
self.selected_label = {} # map selection IDs -> text objects
self.selected_label[1] = self.fig.text(0.6, 0.85, "No structure selected", fontsize=18,
color=self.hub.colors[1])
self.selected_label[2] = self.fig.text(0.6, 0.8, "No structure selected", fontsize=18,
color=self.hub.colors[2])
self.selected_label[3] = self.fig.text(0.6, 0.75, "No structure selected", fontsize=18,
color=self.hub.colors[3])
x = [p.vertices[:, 0] for p in self.lines.get_paths()]
y = [p.vertices[:, 1] for p in self.lines.get_paths()]
xmin = np.min(x)
xmax = np.max(x)
ymin = np.min(y)
ymax = np.max(y)
self.lines.set_picker(2.)
self.lines.set_zorder(0)
dx = xmax - xmin
self.ax_dendrogram.set_xlim(xmin - dx * 0.1, xmax + dx * 0.1)
self.ax_dendrogram.set_ylim(ymin * 0.5, ymax * 2.0)
self.ax_dendrogram.set_yscale('log')
self.fig.canvas.mpl_connect('pick_event', self.line_picker)
self.fig.canvas.mpl_connect('button_press_event', self.select_from_map)
def show(self):
import matplotlib.pyplot as plt
plt.show()
def update_slice(self, pos=None):
if self.array.ndim == 2:
self.image.set_array(self.array)
else:
self.slice = int(round(pos))
self.image.set_array(self.array[self.slice, :, :])
self.update_contours()
self.fig.canvas.draw()
def _connect_to_hub(self):
self.hub.add_callback(self._on_selection_change)
def _on_selection_change(self, selection_id):
self._update_lines(selection_id)
self.update_contours()
self.fig.canvas.draw()
def update_vmin(self, vmin):
if vmin > self._clim[1]:
self._clim = (self._clim[1], self._clim[1])
else:
self._clim = (vmin, self._clim[1])
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def update_vmax(self, vmax):
if vmax < self._clim[0]:
self._clim = (self._clim[0], self._clim[0])
else:
self._clim = (self._clim[0], vmax)
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def select_from_map(self, event):
# Only do this if no tools are currently selected
if event.canvas.toolbar.mode != '':
return
if event.button not in self.selected_label:
return
if event.inaxes is self.ax_image:
input_key = event.button
# Find pixel co-ordinates of click
ix = int(round(event.xdata))
iy = int(round(event.ydata))
if self.array.ndim == 2:
indices = (iy, ix)
else:
indices = (self.slice, iy, ix)
# Select the structure
structure = self.dendrogram.structure_at(indices)
self.hub.select(input_key, structure)
# Re-draw
event.canvas.draw()
def line_picker(self, event):
# Only do this if no tools are currently selected
if event.canvas.toolbar.mode != '':
return
if event.mouseevent.button not in self.selected_label:
return
input_key = event.mouseevent.button
# event.ind gives the indices of the paths that have been selected
# Find levels of selected paths
peaks = [event.artist.structures[i].get_peak(subtree=True)[1] for i in event.ind]
# Find position of minimum level (may be duplicates, let Numpy decide)
ind = event.ind[np.argmax(peaks)]
# Extract structure
structure = event.artist.structures[ind]
# If 3-d, select the slice
if self.slice_slider is not None:
peak_index = structure.get_peak(subtree=True)
self.slice_slider.set_val(peak_index[0][0])
# Select the structure
self.hub.select(input_key, structure)
# Re-draw
event.canvas.draw()
def _update_lines(self, selection_id):
structures = self.hub.selections[selection_id]
select_subtree = self.hub.select_subtree[selection_id]
structure = structures[0]
# Remove previously selected collection
if selection_id in self.selected_lines:
self.ax_dendrogram.collections.remove(self.selected_lines[selection_id])
del self.selected_lines[selection_id]
if structure is None:
self.selected_label[selection_id].set_text("No structure selected")
self.remove_contour(selection_id)
self.fig.canvas.draw()
return
self.remove_all_contours()
if len(structures) <= 1:
label_text = "Selected structure: {0}".format(structure.idx)
elif len(structures) <=3:
label_text = "Selected structures: {0}".format(', '.join([str(structure.idx) for structure in structures]))
else:
label_text = "Selected structures: {0}...".format(', '.join([str(structure.idx) for structure in structures[:3]]))
self.selected_label[selection_id].set_text(label_text)
# Get collection for this substructure
self.selected_lines[selection_id] = self.plotter.get_lines(
structures=structures, subtree=select_subtree)
self.selected_lines[selection_id].set_color(self.hub.colors[selection_id])
self.selected_lines[selection_id].set_linewidth(2)
self.selected_lines[selection_id].set_zorder(structure.height)
# Add to axes
self.ax_dendrogram.add_collection(self.selected_lines[selection_id])
def remove_contour(self, selection_id):
if selection_id in self.selected_contour:
for collection in self.selected_contour[selection_id].collections:
self.ax_image.collections.remove(collection)
del self.selected_contour[selection_id]
def remove_all_contours(self):
""" Remove all selected contours. """
for key in self.selected_contour.keys():
self.remove_contour(key)
def update_contours(self):
self.remove_all_contours()
for selection_id in self.hub.selections.keys():
structures = self.hub.selections[selection_id]
select_subtree = self.hub.select_subtree[selection_id]
struct = structures[0]
if struct is None:
continue
if select_subtree:
mask = struct.get_mask(subtree=True)
else:
mask = reduce(np.add, [structure.get_mask(subtree=True) for structure in structures])
if self.array.ndim == 3:
mask = mask[self.slice, :, :]
self.selected_contour[selection_id] = self.ax_image.contour(
mask, colors=self.hub.colors[selection_id],
linewidths=2, levels=[0.5], alpha=0.75, zorder=struct.height)
class Visualizer:
def __init__(self, field, fieldname, halospec=None):
"""Initializes a visualization instance, that is a windows with a field
field is a 3D numpy array
fieldname is a string with the name of the field
halospec is a 2x2 array with the definition of the halo size
After this call the window is shown
"""
self.field = field
self.fieldname = fieldname
# Register halo information
if halospec is None:
halospec = [[3, 3], [3, 3]]
self.istart = halospec[0][0]
self.iend = field.shape[0] - halospec[0][1]
self.jstart = halospec[1][0]
self.jend = field.shape[1] - halospec[1][1]
self.plotHalo = True
self.plotLogLog = False
self.curklevel = 0
self.figure = plt.figure()
# Slider
slideraxes = plt.axes([0.15, 0.02, 0.5, 0.03], axisbg="lightgoldenrodyellow")
self.slider = Slider(slideraxes, "K level", 0, field.shape[2] - 1, valinit=0)
self.slider.valfmt = "%2d"
self.slider.set_val(0)
self.slider.on_changed(self.updateSlider)
# CheckButton
self.cbaxes = plt.axes([0.8, -0.04, 0.12, 0.15])
self.cbaxes.set_axis_off()
self.cb = CheckButtons(self.cbaxes, ("Halo", "Logscale"), (self.plotHalo, self.plotLogLog))
self.cb.on_clicked(self.updateButton)
# Initial plot
self.fieldaxes = self.figure.add_axes([0.1, 0.15, 0.9, 0.75])
self.collection = plt.pcolor(self._getField(), axes=self.fieldaxes)
self.colorbar = plt.colorbar()
self.fieldaxes.set_xlim(right=self._getField().shape[1])
self.fieldaxes.set_ylim(top=self._getField().shape[0])
plt.xlabel("i")
plt.ylabel("j")
self.title = plt.title("%s - Level 0" % (fieldname,))
plt.show(block=False)
def updateSlider(self, val):
if val == self.curklevel:
return
self.curklevel = round(val)
self.title.set_text("%s - Level %d" % (self.fieldname, self.curklevel))
# Draw new field level
field = self._getField()
size = field.shape[0] * field.shape[1]
array = field.reshape(size)
self.collection.set_array(array)
self.colorbar.set_clim(vmin=field.min(), vmax=field.max())
self.collection.set_clim(vmin=field.min(), vmax=field.max())
self.colorbar.update_normal(self.collection)
self.figure.canvas.draw_idle()
def updateButton(self, label):
if label == "Halo":
self.plotHalo = not self.plotHalo
if label == "Logscale":
self.plotLogLog = not self.plotLogLog
self.updatePlot()
def updatePlot(self):
# Redraw field
self.collection.remove()
field = self._getField()
if self.plotLogLog:
minvalue = field.min()
norm = SymLogNorm(linthresh=1e-10)
self.collection = plt.pcolor(field, axes=self.fieldaxes, norm=norm)
self.colorbar.set_clim(vmin=minvalue, vmax=field.max())
else:
self.collection = plt.pcolor(field, axes=self.fieldaxes)
self.colorbar.set_clim(vmin=field.min(), vmax=field.max())
self.colorbar.set_norm(norm=Normalize(vmin=field.min(), vmax=field.max()))
self.fieldaxes.set_xlim(right=field.shape[1])
self.fieldaxes.set_ylim(top=field.shape[0])
self.colorbar.update_normal(self.collection)
self.figure.canvas.draw_idle()
def _getField(self):
if self.plotHalo:
return np.rot90(self.field[:, :, self.curklevel])
else:
return np.rot90(self.field[self.istart : self.iend, self.jstart : self.jend, self.curklevel])
def plot_biplot(x, information, rhomin, rhomax):
## Formatting input
if callable(information):
n = 100
width = 20
# x = np.linspace(0, 1, n)
y = information(x)
else:
n = len(information)
width = int(n/5)
# x = np.arange(n)
y = information
array = np.repeat(y, width).reshape((len(y), width))
# Font type for all plots
rc('font',**{'family':'serif','serif':['Palation'], 'size':24, 'weight':'bold'})
rc('text', usetex=True)
mpl.rcParams['text.latex.preamble']=[r"\usepackage{amsmath}"]
# Creating plot
# fig = plt.figure(figsize=(18,10))
fig1, ax1 = plt.subplots(figsize=(2, 10))
# gs = gridspec.GridSpec(8, 4)
# ax1 = fig.add_subplot(gs[1:, 0])
# ax1 = fig.add_axes([0.05, 0.05, 0.05, 0.8])
# ax2 = fig.add_axes([0.2, 0.05, 0.75, 0.8])
#ax1 = fig.add_subplot(111)
#ax2 = fig.add_subplot(111)
colormap_chosen = mpl.colors.LinearSegmentedColormap.from_list('mycolors',['red','yellow']) #['yellow','yellow','#F7F8E0','yellow','red','#8A0808']
colormap_chosen = 'OrRd_r'#'gist_heat' #hot
cm = plt.get_cmap(colormap_chosen)
cNorm = colors.Normalize(vmin=rhomin, vmax=rhomax)
ax1.imshow(array, cmap=cm, norm=cNorm)
# ax2 = fig.add_subplot(gs[1:, 1:5])
fig1 = plt.gcf()
fig2, ax2 = plt.subplots(figsize=(10, 8))
x_0, rho_0 = evolver.rho_aprox(0.5)
ax2.plot(rho_0, x_0, linewidth=5, linestyle="-", c="grey", zorder=1, alpha=0.3)
ax2.plot(information, x, linewidth=8, linestyle="-", c="black", zorder=1)
ax2.plot(information, x, linewidth=5, linestyle="-", c="white", zorder=1)
##### TRYING TO MAKE LINES WITH COLOR GRADIENTS
# ax2.scatter(information,x,c=range(len(information)), marker='_', s=30)
# path = mpath.Path(np.column_stack([y, x]))
# verts = path.interpolated(steps=3).vertices
# xcolor, ycolor = verts[:, 0], verts[:, 1]
colorline(ax2, y, x, z=rhomin*(rhomin-y)/(rhomin-rhomax) + rhomax*(rhomax-y)/(rhomax-rhomin), cmap=cm, norm=cNorm, linewidth=5, alpha=1)
# gs.update(wspace=0.5, hspace=0.5)
# Stetic tuning of plot
ax2.set_ylim([min(x),max(x)])
ax2.set_xlim([990,1093])
ax2.set_xlabel(r"\textbf{Density / g L}$\boldsymbol{^{-1}}$")
ax2.set_ylabel(r"\textbf{Height / mm}",rotation=270, labelpad= 10)#fontsize=20,
ax2.yaxis.set_label_position("right")
ax2.tick_params(labeltop=True)
ax2.grid(True)
plt.setp(ax2.get_xticklabels(), fontsize=18)
# Final Configuration
plt.setp(ax1.get_xticklabels(), visible=False)
plt.setp(ax1.get_yticklabels(), visible=False)
plt.setp(ax2.get_yticklabels(), visible=False)
ax1.get_xaxis().set_visible(False)
for axis in ['top','bottom','left','right']:
ax1.spines[axis].set_linewidth(5)
# fig = plt.gcf()
fig2 = plt.gcf()
fig3, axslider = plt.subplots(figsize=(10, 1))
#axslider = plt.axes([0.3, 0.94, 0.5, 0.04], axisbg=None)
# axslider = fig.add_subplot(gs[0,1:3])
samp = Slider(axslider, 'Time (min)', 0., maxtime, valinit=0,color='grey',alpha=0.3,valfmt='%i'.ljust(5))
samp.set_val(float(time)/60.)
fig3 = plt.gcf()
return fig1, fig2, fig3
class py3DSeedEditor:
""" Viewer and seed editor for 2D and 3D data.
py3DSeedEditor(img, ...)
img: 2D or 3D grayscale data
voxelsizemm: size of voxel, default is [1, 1, 1]
initslice: 0
colorbar: True/False, default is True
cmap: colormap
zaxis: axis with slice numbers
ed = py3DSeedEditor(img)
ed.show()
selected_seeds = ed.seeds
"""
def __init__(self, img, voxelsizemm=[1,1,1], initslice = 0 , colorbar = True,
cmap = matplotlib.cm.Greys_r, seeds = None, contour = None, zaxis=0,
mouse_button_map= {1:1, 2:2, 3:3, 4:4, 5:5, 6:6, 7:7, 8:8},
windowW = [], windowC = [],
range_per_slice = False
):
self.fig = plt.figure()
if len(img.shape) == 2:
imgtmp = img
img = np.zeros([1, imgtmp.shape[0], imgtmp.shape[1]])
img[-1,:,:] = imgtmp
zaxis = 0
# Rotate data in depndecy on zaxispyplot
img = self._rotate_start(img, zaxis)
seeds = self._rotate_start(seeds, zaxis)
contour = self._rotate_start(contour, zaxis)
self.rotated_back = False
self.zaxis = zaxis
# if True, intensity range is calculated per slice = better visualisation for
# higher number of labels
self.range_per_slice = range_per_slice
#self.ax = self.fig.add_subplot(111)
self.imgshape = list(img.shape)
self.img = img
self.actual_slice = initslice
self.colorbar = colorbar
self.cmap = cmap
if seeds == None:
self.seeds = np.zeros(self.imgshape, np.int8)
else:
self.seeds = seeds
if not (windowW and windowC):
self.imgmax = np.max(img)
self.imgmin = np.min(img)
else:
self.imgmax = windowC + (windowW / 2)
self.imgmin = windowC - (windowW / 2)
""" Mapping mouse button to class number. Default is normal order"""
self.button_map = mouse_button_map
self.contour = contour
self.press = None
self.press2 = None
# language
self.texts = {'btn_delete':'Delete', 'btn_close': 'Close'}
#iself.fig.subplots_adjust(left=0.25, bottom=0.25)
self.ax = self.fig.add_axes([0.2, 0.3, 0.7,0.6])
self.draw_slice()
if self.colorbar:
self.fig.colorbar(self.imsh)
# user interface look
axcolor = 'lightgoldenrodyellow'
ax_actual_slice = self.fig.add_axes([0.2, 0.2, 0.6, 0.03], axisbg=axcolor)
self.actual_slice_slider = Slider(ax_actual_slice, 'Slice', 0,
self.imgshape[2], valinit=initslice)
# conenction to wheel events
self.fig.canvas.mpl_connect('scroll_event', self.on_scroll)
self.actual_slice_slider.on_changed(self.sliceslider_update)
# draw
self.fig.canvas.mpl_connect('button_press_event', self.on_press)
self.fig.canvas.mpl_connect('button_release_event', self.on_release)
self.fig.canvas.mpl_connect('motion_notify_event', self.on_motion)
# delete seeds
self.ax_delete_seeds = self.fig.add_axes([0.2,0.1,0.1,0.075])
self.btn_delete = Button(self.ax_delete_seeds, self.texts['btn_delete'])
self.btn_delete.on_clicked(self.callback_delete)
# close button
self.ax_delete_seeds = self.fig.add_axes([0.7,0.1,0.1,0.075])
self.btn_delete = Button(self.ax_delete_seeds, self.texts['btn_close'])
self.btn_delete.on_clicked(self.callback_close)
self.draw_slice()
def _rotate_start(self, data, zaxis):
if data != None:
if zaxis == 0:
data = np.transpose(data,(1,2,0))
elif zaxis == 2:
pass
else:
print "problem with zaxis in _rotate_start()"
return data
def _rotate_end(self, data, zaxis):
if data != None:
if self.rotated_back == False:
if zaxis == 0:
data = np.transpose(data,(2,0,1))
elif zaxis == 2:
pass
else:
print "problem with zaxis in _rotate_start()"
else:
print "There is a danger in calling show() twice"
return data
def update_slice(self):
#TODO tohle je tu kvuli contour, neumim ji odstranit jinak
self.ax.cla()
self.draw_slice()
def draw_slice(self):
self.actual_slice = np.int(self.actual_slice)
sliceimg = self.img[:, :, self.actual_slice]
if self.range_per_slice:
self.imsh = self.ax.imshow(sliceimg, self.cmap, vmin=sliceimg.min(), vmax=sliceimg.max(), interpolation='nearest')
else:
self.imsh = self.ax.imshow(sliceimg, self.cmap, vmin=self.imgmin, vmax=self.imgmax, interpolation='nearest')
self.ax.imshow(self.prepare_overlay(self.seeds[:, :, self.actual_slice]), interpolation='nearest', vmin=self.imgmin, vmax=self.imgmax)
if self.contour != None:
try:
# exception catch problem with none object in image
ctr = self.ax.contour(self.contour[:, :, self.actual_slice], 1, linewidths=2)
except:
pass
self.fig.canvas.draw()
def next_slice(self):
self.actual_slice = self.actual_slice + 1
if self.actual_slice >= self.imgshape[2]:
self.actual_slice = 0
def prev_slice(self):
self.actual_slice = self.actual_slice - 1
if self.actual_slice < 0:
self.actual_slice = self.imgshape[2] - 1
def sliceslider_update(self, val):
# zaokrouhlení
#self.actual_slice_slider.set_val(round(self.actual_slice_slider.val))
self.actual_slice = round(val)
self.update_slice()
def prepare_overlay(self,seeds):
sh = list(seeds.shape)
if len(sh) == 2:
sh.append(4)
else:
sh[2] = 4
overlay = np.zeros(sh)
overlay[:,:,0] = (seeds == 1)
overlay[:,:,1] = (seeds == 2)
overlay[:,:,2] = (seeds == 3)
overlay[:,:,3] = (seeds > 0)
return overlay
def show(self):
""" Function run viewer window.
"""
plt.show()
# Rotate data in depndecy on zaxis
self.img = self._rotate_end(self.img, self.zaxis)
self.seeds = self._rotate_end(self.seeds, self.zaxis)
self.contour = self._rotate_end(self.contour, self.zaxis)
self.rotated_back = True
return self.seeds
def on_scroll(self, event):
''' mouse wheel is used for setting slider value'''
if event.button == 'up':
self.next_slice()
if event.button == 'down':
self.prev_slice()
self.actual_slice_slider.set_val (self.actual_slice)
## malování -------------------
def on_press(self, event):
'on but-ton press we will see if the mouse is over us and store some data'
if event.inaxes != self.ax: return
#contains, attrd = self.rect.contains(event)
#if not contains: return
#print 'event contains', self.rect.xy
#x0, y0 = self.rect.xy
self.press = [event.xdata], [event.ydata], event.button
#self.press1 = True
def on_motion(self, event):
'on motion we will move the rect if the mouse is over us'
if self.press is None: return
if event.inaxes != self.ax: return
#print event.inaxes
x0, y0, btn = self.press
x0.append(event.xdata)
y0.append(event.ydata)
def on_release(self, event):
'on release we reset the press data'
if self.press is None: return
#print self.press
x0, y0, btn = self.press
if btn == 1:
color = 'r'
elif btn == 2:
color = 'b'
#button Mapping
btn = self.button_map[btn]
self.set_seeds(y0, x0, self.actual_slice, btn )
self.press = None
self.update_slice()
def callback_delete(self, event):
self.seeds[:,:,self.actual_slice] = 0
self.update_slice()
def callback_close(self, event):
matplotlib.pyplot.clf()
matplotlib.pyplot.close()
def set_seeds(self, px, py, pz, value = 1, voxelsizemm = [1,1,1], cursorsizemm = [1,1,1]):
assert len(px) == len(py) , 'px and py describes a point, their size must be same'
for i, item in enumerate(px):
self.seeds[item, py[i], pz] = value
def get_seed_sub(self, label):
""" Return list of all seeds with specific label
"""
sx, sy, sz = np.nonzero(self.seeds == label)
return sx, sy, sz
def get_seed_val(self, label):
""" Return data values for specific seed label"""
return self.img[self.seeds==label]
class Parameters(object):
"""
An interactive matplotlib window to test and set paramters for eye tracking
on a subset of frames.
"""
def __init__(self, notes):
self.wintitle = notes.wintitle
self.stack = notes.stack # cropped frame stack
self.frame_dims = notes.stack.shape[-2:] # cropped frame dimensions
self.nframes = notes.stack.shape[0] # number of frames
_, _, _, mmperpx = geom.parse_axis(notes.axis) # mm to pixel conversion factor
self.diffs = notes.diffs # mean pixel-wise differences from reference frame
# status variable, set to True on 'c' keypress to exit parameter-setting
# loop and perform eye tracking on all files:
self.done = False
# define a parameter dictionary with some initial values
# first specify maximum/ initial values for some of the parameters
method = 'convolve' # eye tracking method ('threshold' or 'convolve')
shape = 'lse_ellipse' # shape fitting methos ('lse_ellipse or 'min_enclosing')
# parameters for image pre-processing:
eq_sp_maxval = 1 # bihist equalization separation point
eq_rp_maxval = 1 # bihist equalization range point
c_sig_maxval = 255 / 2 # intensity space sigma for bilateral filter
s_sig_maxval = max(self.frame_dims) / 8 # spatial sigma for bilateral filter
k_size_maxval = max(self.frame_dims) / 8 # kernel size for median filter
k_size_inival = np.int(np.round(k_size_maxval/2))
if k_size_inival % 2 == 0: # must be an odd integer
k_size_inival += 1
# parameters for threshold method:
dark_thr_inival = 255 * 0.25 # threshold for dark areas
light_thr_inival = 255 * 0.75 # threshold for light areas
area_thr_maxval = np.pi * (0.5 / mmperpx)**2 # area threshold
# parameters for convolve method:
conv_size_maxval = 0.5 / mmperpx # kernel size for convolution
conv_size_inival = np.int(np.round(conv_size_maxval/4))
if conv_size_inival % 2 == 0: # must be an odd integer
conv_size_inival += 1
rad_maxval = np.round(2/mmperpx).astype('int') # max radius for edges
# parameters for blink detection
b_thr_inival = np.median(self.diffs)*3
# build parameter dictionary
self.params = {'eq':False, 'eq_sp':0, 'eq_rp':0, 'k_size':k_size_inival,
'c_sig':c_sig_maxval/2, 's_sig':s_sig_maxval/2,
'dark_thr':20, 'light_thr':235, 'area_thr':0,
'conv_size':conv_size_inival, 'max_rad':rad_maxval/2,
'shape':shape, 'ht_fit':False, 'method':method,
'blink_thr':b_thr_inival,}
# set initial frames to display
self.frame_ind = 0 # current frame index
self.frame = self.stack[self.frame_ind,:,:] # original frame
# processed (filtered and equalized) frame:
self.p_frame = img.pre_process(self.frame, self.params)
# inverted binarized frame (dark contours):
self.b_frame, _ = img.binarize(self.p_frame, self.params)
# p_frame convolved with a black square (center = argmin[c_frame])
self.c_frame, self.center = img.square_convolve(self.p_frame, self.params['k_size'])
self.g_frame = img.gradient(self.p_frame) # gradient magnitude of p_frame
self.edge_pts = geom.starburst(self.g_frame, self.center, self.params['max_rad'], 100)
# set up figure with plots and images
self.fig, _ = plt.subplots()
self.fig.canvas.set_window_title(self.wintitle)
grid = gs.GridSpec(12, 12)
# original frame display axis
self.ax_frame = plt.subplot(grid[:6, :4])
self.ax_frame.axis('off')
# processed frame display axis
self.ax_pframe = plt.subplot(grid[6:9, :2])
self.ax_pframe.axis('off')
self.ax_pframe.set_title('Processed Frame')
# dark binary display axis
self.ax_bframe = plt.subplot(grid[6:9, 2:4])
self.ax_bframe.axis('off')
self.ax_bframe.set_title('Dark Contours')
# convolution display axis
self.ax_cframe = plt.subplot(grid[9:12, :2])
self.ax_cframe.axis('off')
self.ax_cframe.set_title('Convolution')
self.c_dot = self.ax_cframe.scatter(self.center[0], self.center[1], s=8)
# gradient display axis
self.ax_gframe = plt.subplot(grid[9:12, 2:4])
self.ax_gframe.axis('off')
self.ax_gframe.set_title('Gradient')
self.e_dots = self.ax_gframe.scatter(self.edge_pts[:,0], self.edge_pts[:,1],
s=1, color='C0')
# intensity histogram
self.ax_hist = plt.subplot(grid[:4, 4:7])
self.hist = self.ax_hist.hist(self.p_frame.ravel(), bins=np.arange(256),
color='C0', alpha=0.5, density=True)
self.dark_thr = self.ax_hist.axvline(self.params['dark_thr'], color=[0,0,0],
marker=' ', label='Dark Threshold')
self.light_thr = self.ax_hist.axvline(self.params['light_thr'], color=[.75,.75,.75],
marker=' ', label='Light Threshold')
self.ax_hist.set_title('Intensity Histogram')
self.ax_hist.legend()
# timecourse of mean differences between each frame and the reference
self.ax_diffs = plt.subplot(grid[4:8, 4:7])
self.ax_diffs.scatter(np.linspace(0,len(self.diffs),num=len(self.diffs)),
self.diffs, s=3, color='C0')
self.ax_diffs.set_xlim([0, self.nframes])
self.f_dot = self.ax_diffs.scatter(0, self.diffs[0], s=4, color='C1')
self.blink_thr1 = self.ax_diffs.axhline(self.params['blink_thr'], marker=' ',
color=[0,0,0], label='Blink Threshold')
self.ax_diffs.legend()
self.ax_diffs.set_ylabel('Diff. from reference')
# slider for frame scrolling
self.ax_fslider = plt.subplot(grid[8, 4:7])
self.f_slider = Slider(self.ax_fslider, 'Frame', 0, self.nframes-1,
valinit=0, valfmt='%d')
# distribution of the differences
self.ax_dist = plt.subplot(grid[9:12, 4:7])
self.ax_dist.hist(self.diffs.ravel(), bins=100, color='C0', density=True)
self.blink_thr2 = self.ax_dist.axvline(self.params['blink_thr'], marker=' ',
color=[0,0,0])
self.ax_dist.set_xlabel('Diff. from reference')
# slider for eq_sp equalization parameter
self.ax_spslider = plt.subplot(grid[0, 8:11])
self.sp_slider = Slider(self.ax_spslider, 'EQ Sep. Point', 0, 1,
valinit=self.params['eq_sp'], valfmt='%02f')
# slider for eq_rp equalization parameter
self.ax_rpslider = plt.subplot(grid[1, 8:11])
self.rp_slider = Slider(self.ax_rpslider, 'EQ Range Point', 0, 1,
valinit=self.params['eq_rp'], valfmt='%02f')
# slider for c_sig filter parameter
self.ax_cslider = plt.subplot(grid[3, 8:11])
self.c_slider = Slider(self.ax_cslider, 'Color Sigma', 0, c_sig_maxval,
valinit=self.params['c_sig'], valfmt='%d')
# slider for s_sig filter parameter
self.ax_sslider = plt.subplot(grid[4, 8:11])
self.s_slider = Slider(self.ax_sslider, 'Spatial Sigma', 0, s_sig_maxval,
valinit=self.params['s_sig'], valfmt='%d')
# slider for k_size filter parameter
self.ax_kslider = plt.subplot(grid[5, 8:11])
self.k_slider = Slider(self.ax_kslider, 'Kernel Size', 1, k_size_maxval,
valinit=k_size_inival, valfmt='%d')
# slider for area_thr parameter
self.ax_aslider = plt.subplot(grid[7, 8:11])
self.a_slider = Slider(self.ax_aslider, 'Area Thresh', 0, area_thr_maxval,
valinit=0, valfmt='%d')
# slider for conv_size convolution parameter
self.ax_ckslider = plt.subplot(grid[9, 8:11])
self.ck_slider = Slider(self.ax_ckslider, 'Conv. Kernel Size', 0, conv_size_maxval,
valinit=self.params['conv_size'], valfmt='%d')
# slider for max_rad starburst parameter
self.ax_rslider = plt.subplot(grid[10, 8:11])
self.r_slider = Slider(self.ax_rslider, 'Max. Radius', 0, rad_maxval,
valinit=self.params['max_rad'], valfmt='%d')
# disconnect default matplotlib key bindings
manager, canvas = self.fig.canvas.manager, self.fig.canvas
canvas.mpl_disconnect(manager.key_press_handler_id)
# maximize display window
manager.window.showMaximized()
# connect callback functions
self.cidpress = self.fig.canvas.mpl_connect('button_press_event', self.on_click)
self.cidkey = self.fig.canvas.mpl_connect('key_press_event', self.on_key)
self.f_slider.on_changed(self.update_frame)
self.sp_slider.on_changed(self.sp_update)
self.rp_slider.on_changed(self.rp_update)
self.c_slider.on_changed(self.c_update)
self.s_slider.on_changed(self.s_update)
self.k_slider.on_changed(self.k_update)
self.a_slider.on_changed(self.a_update)
self.ck_slider.on_changed(self.ck_update)
self.r_slider.on_changed(self.r_update)
# get pupil and reflection patches
self.get_eye_patches()
# display
self.update_display()
# display user prompts
print("Set Parameters")
print(" - click on the slider or use the arrow keys to scroll between video frames")
print("")
print(" Pre-Processing:")
print(" - use 'e' to toggle histogram equalization")
print(" - set parameters for equalizationg and smoothing using the sliders")
print("")
print(" - use 'm' to change eye tracking method")
print(" Threshold Method:")
print(" - set dark and light binarization thresholds by clicking on the")
print(" intensity histogram (right & left clicks respectively)")
print(" - set an area threshold using the slider")
print(" Convolution Method:")
print(" - set a kernel size and a maximum radius using the sliders")
print("")
print(" Ellipse Fitting:")
print(" - use 'h' to toggle ellipse fitting via re-sampling (Hough Transform)")
print("")
print(" - set the blink-detection threshold by clicking on the difference")
print(" scatter plot or distribution plot")
print("")
print(" - use 'r' to re-set paramters and return to the previous window,")
print(" 'c' to confirm the current parameters and commence tracking,")
print(" or 'esc' to quit (parameters will not be saved)")
def on_click(self, event):
"""sets thresholds based on click"""
# set binarization thresholds
if event.inaxes == self.ax_hist:
if event.button == 1: # left click
self.dark_thr.remove()
self.params['dark_thr'] = event.xdata
self.dark_thr = self.ax_hist.axvline(self.params['dark_thr'],
color=[0,0,0], marker=' ',
label='Dark Threshold')
elif event.button == 3: # right click
self.light_thr.remove()
self.params['light_thr'] = event.xdata
self.light_thr = self.ax_hist.axvline(self.params['light_thr'],
color=[0.75,0.75,0.75], marker=' ',
label='Light Threshold')
# update binarized frame
self.b_frame, light = img.binarize(self.p_frame, self.params)
# update eye patches
self.get_eye_patches()
# set blink threshold in diffs timecourse axis
elif event.inaxes == self.ax_diffs:
if event.button == 1:
self.blink_thr1.remove()
self.blink_thr2.remove()
self.params['blink_thr'] = event.ydata
self.blink_thr1 = self.ax_diffs.axhline(self.params['blink_thr'],
marker=' ', color=[0,0,0],
label='Blink Threshold')
self.blink_thr2 = self.ax_dist.axvline(self.params['blink_thr'],
marker=' ', color=[0,0,0])
# set blink threshold in diffs distribution axis
elif event.inaxes == self.ax_dist:
if event.button == 1:
self.blink_thr1.remove()
self.blink_thr2.remove()
self.params['blink_thr'] = event.xdata
self.blink_thr1 = self.ax_diffs.axhline(self.params['blink_thr'],
marker=' ', color=[0,0,0],
label='Blink Threshold')
self.blink_thr2 = self.ax_dist.axvline(self.params['blink_thr'],
marker=' ', color=[0,0,0])
# update display
self.update_display()
def on_key(self, event):
"""specifies functions of pressed keys in matplotlib figure"""
# toggle histogram equalization
if event.key == 'e':
self.params['eq'] = not self.params['eq']
print(" EQ: " + str(self.params['eq']))
self.update_pframes()
self.update_hist()
self.get_eye_patches()
self.update_display()
# toggle eye tracking method
elif event.key == 'm':
if self.params['method'] == 'threshold':
self.params['method'] = 'convolve'
print(" Method: convolve")
elif self.params['method'] == 'convolve':
self.params['method'] = 'threshold'
print(" Method: threshold")
self.get_eye_patches() # update eye patches
self.update_display()
# toggle ellipse fitting method
elif event.key == 'f':
if self.params['shape'] == 'lse_ellipse':
self.params['shape'] = 'min_enclosing'
elif self.params['shape'] == 'min_enclosing':
self.params['shape'] = 'lse_ellipse'
print(" Shape fit: " + self.params['shape'])
self.get_eye_patches()
self.update_display()
# toggle Hough transform ellipse fitting
elif event.key == 'h':
self.params['ht_fit'] = not self.params['ht_fit']
print(" Hough transform: " + str(self.params['ht_fit']))
self.get_eye_patches() # update eye patches
self.update_display()
# change frame
elif event.key == 'right': # move forward one frame
if np.round(self.f_slider.val) < self.nframes-1:
self.f_slider.set_val(self.f_slider.val+1)
elif event.key == 'left': # move back one frame
if np.round(self.f_slider.val) > 0:
self.f_slider.set_val(self.f_slider.val-1)
# continue, close, or re-set
elif event.key == 'c': # confirm parameters and close figure
print("Eye tracking parameters confirmed")
self.done = True
plt.close('all')
elif event.key == 'escape': # quit eye tracking
print("Eye tracking aborted")
plt.close('all')
sys.exit()
elif event.key == 'r': # re-set & return to annotation
print("Parameters re-set")
plt.close('all')
def update_frame(self, val):
"""updates frame and plots based on frame slider value"""
# update frame index
self.frame_ind = np.int(np.round(self.f_slider.val))
# update displays
self.frame = self.stack[self.frame_ind,:,:] # raw frame
self.update_pframes()
self.update_hist()
# diffs plot
self.f_dot.remove()
self.f_dot = self.ax_diffs.scatter(self.frame_ind, self.diffs[self.frame_ind],
s=4, color=[1,0,1])
self.get_eye_patches()
self.update_display()
def sp_update(self, val):
"""updates eq_sp equalization parameter and re-applies filters"""
self.params['eq_sp'] = self.sp_slider.val
self.update_pframes()
self.update_hist()
self.get_eye_patches()
self.update_display()
def rp_update(self, val):
"""updates eq_rp equalization parameter and re-applies filters"""
self.params['eq_rp'] = self.rp_slider.val
self.update_pframes()
self.update_hist()
self.get_eye_patches()
self.update_display()
def c_update(self, val):
"""updates c_sig filter parameter and re-applies filters"""
self.params['c_sig'] = self.c_slider.val
self.update_pframes()
self.get_eye_patches()
self.update_display()
def s_update(self, val):
"""updates s_sig filter parameter and re-applies filters"""
self.params['s_sig'] = self.s_slider.val
self.update_pframes()
self.get_eye_patches()
self.update_display()
def k_update(self, val):
"""updates k_size filter parameter and re-applies filters"""
self.params['k_size'] = np.int(np.round(self.k_slider.val))
if self.params['k_size'] % 2 == 0: # kernel size must be odd
self.params['k_size'] += 1
self.update_pframes()
self.get_eye_patches()
self.update_display()
def a_update(self, val):
"""updates dark contour area threshold and re-applies eye-tracking"""
self.params['area_thr'] = self.a_slider.val
self.get_eye_patches()
self.update_display()
def ck_update(self, val):
"""updates conv_size convolution parameter and re-applies eye-tracking"""
self.params['conv_size'] = np.int(np.round(self.ck_slider.val))
if self.params['conv_size'] % 2 == 0: # kernel size must be odd
self.params['conv_size'] += 1
self.update_pframes()
self.get_eye_patches()
self.update_display()
def r_update(self, val):
"""updates max_rad and re-applies eye-tracking"""
self.params['max_rad'] = self.r_slider.val
self.update_pframes()
self.get_eye_patches()
self.update_display()
def update_pframes(self):
"""updates the filtered frame"""
self.p_frame = img.pre_process(self.frame, self.params)
self.b_frame, light = img.binarize(self.p_frame, self.params)
self.c_frame, self.center = img.square_convolve(self.p_frame, self.params['conv_size'])
self.c_dot.remove()
self.c_dot = self.ax_cframe.scatter(self.center[0], self.center[1], s=8, c='C0')
self.g_frame = img.gradient(self.p_frame)
self.edge_pts = geom.starburst(self.g_frame, self.center, self.params['max_rad'], 100)
self.e_dots.remove()
self.e_dots = self.ax_gframe.scatter(self.edge_pts[:,0], self.edge_pts[:,1],
s=1, color='C0')
def update_hist(self):
"""updates the intensity histogram"""
# remove old histogram patches
_ = [i.remove() for i in self.hist[2]]
# update histogram
self.hist = self.ax_hist.hist(self.p_frame.ravel(), bins=np.arange(256),
color='C0', alpha=0.5, density=True)
# set ylim to match current histogram
self.ax_hist.set_ylim(top=self.hist[0].max())
def update_display(self):
"""updates the display"""
self.ax_frame.imshow(self.frame, cmap='gray')
self.ax_pframe.imshow(self.p_frame, cmap='gray')
self.ax_bframe.imshow(self.b_frame, cmap='gray')
self.ax_cframe.imshow(self.c_frame, cmap='gray')
self.ax_gframe.imshow(self.g_frame, cmap='gray')
self.fig.canvas.draw()
def get_eye_patches(self):
"""performs eye tracking on the current frame and updates pupil and
reflection patches"""
if self.params['method'] == 'threshold':
pupil, cr = main.itrack_threshold(self.frame, self.params)
elif self.params['method'] == 'convolve':
pupil, cr = main.itrack_convolve(self.frame, self.params)
# clear any old patches
self.ax_frame.patches.clear()
self.ax_pframe.patches.clear()
if not np.isnan(pupil).all(): # if fit was successful
# parse ellipse parameters
xy = (pupil[0], pupil[1])
width, height, theta = pupil[2] * 2, pupil[3] * 2, pupil[4]
# create patches
self.pupil = Ellipse(xy, width, height, angle=theta,
lw=2, ec='C0', fill=False)
self.pupil2 = Ellipse(xy, width, height, angle=theta,
lw=2, ec='C0', fill=False)
# add patches to appropriate axes
self.ax_frame.add_patch(self.pupil)
self.ax_pframe.add_patch(self.pupil2)
# check that tracking was successful
if not np.isnan(cr).all(): # if fit was successful
# parse ellipse parameters
xy = (cr[0], cr[1])
width, height, theta = cr[2] * 2, cr[3] * 3, cr[4]
# create patches
self.cr = Ellipse(xy, width, height, angle=theta,
lw=2, ec='C1', fill=False)
self.cr2 = Ellipse(xy, width, height, angle=theta,
lw=2, ec='C1', fill=False)
# add patches to appropriate axes
self.ax_frame.add_patch(self.cr)
self.ax_pframe.add_patch(self.cr2)
class CompareAnimation:
"""
Launch two parallel animations of runs, so the user can
easily compare the structures. Example:
R1 = ReadRun("fake/path/run_1/")
R2 = ReadRun("fake/path/run_2/")
new_anim=CompareAnimation(R1.S,R2.S)
new_anim.launch()
plt.show()
"""
def __init__(self,snaplist1,snaplist2, symbol="bo", dt = None, markersize=2, **kwargs):
if snaplist1[0].t < snaplist2[0].t:
S = deepcopy(snaplist1[0])
S.t = snaplist2[0].t
snaplist1.reverse()
snaplist1.append(S)
snaplist1.reverse()
if snaplist2[0].t < snaplist1[0].t:
S = deepcopy(snaplist2[0])
S.t = snaplist1[0].t
snaplist2.reverse()
snaplist2.append(S)
snaplist2.reverse()
self.snaplists = [ snaplist1, snaplist2 ]
self.times = [ [s.t for s in snapl] for snapl in self.snaplists ]
#self.nsnap=len(snaplist)
self.n = [ 0, 0 ]
self.t=0.
self.tmax = max(self.snaplists[0][-1].t, self.snaplists[1][-1].t)
self.dt = self.tmax/200. if dt is None else dt
self.delay=1
self.symbol= symbol
self.markersize= markersize
self.pause_switch=True
self.BackgroundColor='white'
self.kwargs=kwargs
def create_frame(self):
self.fig = plt.figure(figsize=(15,10))
print "fig created"
self.ax = []
for i in range(2):
ax = self.fig.add_subplot(121+i, projection='3d',
adjustable='box',
axisbg=self.BackgroundColor)
ax.set_aspect('equal')
plt.tight_layout()
ax.set_xlabel("x (pc)")
ax.set_ylabel ("y (pc)")
ax.set_zlabel ("z (pc)")
self.ax.append(ax)
X,Y,Z = [],[],[]
for snapl in self.snaplists:
X.append( snapl[0].x )
Y.append( snapl[0].y )
Z.append( snapl[0].z )
max_range = np.array([X[0].max()-X[0].min(),
Y[0].max()-Y[0].min(),
Z[0].max()-Z[0].min()]).max() / 3.0
mean_x = X[0].mean(); mean_y = Y[0].mean(); mean_z = Z[0].mean()
self.fig.subplots_adjust(bottom=0.06)#, left=0.1)
self.line, self.canvas = [],[]
for (x,y,z,ax) in zip(X,Y,Z,self.ax):
self.line.append(ax.plot(x, y, z, self.symbol, markersize=self.markersize,
**self.kwargs )[0])
self.canvas.append(ax.figure.canvas)
ax.set_xlim(mean_x - max_range, mean_x + max_range)
ax.set_ylim(mean_y - max_range, mean_y + max_range)
ax.set_zlim(mean_z - max_range, mean_z + max_range)
ax_pauseB=plt.axes([0.04, 0.02, 0.06, 0.025])
self.pauseB=Button(ax_pauseB,'Play')
self.pauseB.on_clicked(self.Pause_button)
slider_ax = plt.axes([0.18, 0.02, 0.73, 0.025])
self.slider_time = Slider(slider_ax,
"Time",
self.snaplists[0][0].t,
self.snaplists[0][-1].t,
valinit = self.snaplists[0][0].t,
color = '#AAAAAA')
self.slider_time.on_changed(self.slider_time_update)
def update_lines(self):
nsnaps = [0,0]
for j,snaplist,time in zip(range(2),self.snaplists,self.times):
ind = np.nonzero(time < self.t)[0]
nsnaps[j] = max(ind) if len(ind)!=0 else 0
for (line,snaplist,n) in zip(self.line, self.snaplists, nsnaps):
line.set_data(snaplist[n].x, snaplist[n].y)
line.set_3d_properties(snaplist[n].z)
for canv in self.canvas:
canv.draw()
def timer_update(self,lines):
if not self.pause_switch:
self.t = self.t+self.dt
if self.t > self.tmax:
self.t = self.t - self.tmax
self.update_lines()
self.slider_time.set_val(self.t)
def Pause_button(self,event):
self.pause_switch = not self.pause_switch
def slider_time_update(self,val):
self.t = val
self.update_lines()
def launch(self):
self.create_frame()
self.timer=self.fig.canvas.new_timer(interval=self.delay)
args=[self.line]
# We tell the timer to call the update function every 100ms
self.timer.add_callback(self.timer_update,*args)
self.timer.start()
class UI:
def __init__(self, ss_plotdata, min_limit, max_limit, coloring, ppf):
self.ss_plotdata = ss_plotdata
self.n = self.ss_plotdata.n
self.ppf = ppf
self.min_limit = min_limit
self.max_limit = max_limit
self.azimuth = -65
self.elevation = 23
self.fig = plt.figure()
self.fig.set_size_inches(17, 9)
gs = gridspec.GridSpec(2,
3,
width_ratios=[1, 1, 0.1],
height_ratios=[1, 0.05],
left=0.05)
self.fig.suptitle(ss_plotdata.title)
self.entry_ax = plt.subplot(gs[0], projection='3d')
self.entry_ax.view_init(elev=self.elevation, azim=self.azimuth)
self.exit_ax = plt.subplot(gs[1], projection='3d')
self.exit_ax.view_init(elev=self.elevation, azim=self.azimuth)
# setup colorbar
self.cb_ax = plt.subplot(gs[2])
if coloring == Coloring.DISCRETE_MONTHS or coloring == Coloring.DISCRETE_MONTHS_SPLIT_MARKERS or coloring == Coloring.DISCRETE_MONTHS_POLYGONS:
cbar = self.fig.colorbar(ss_plotdata.mapping,
cax=self.cb_ax,
label=ss_plotdata.colorbar_label,
ticks=np.arange(1.5, 13.5, 1))
cbar.ax.set_yticklabels(calendar.month_abbr[1:])
else:
self.fig.colorbar(ss_plotdata.mapping,
cax=self.cb_ax,
label=ss_plotdata.colorbar_label)
self.slider_ax = plt.subplot(gs[1, :])
self.time_slider = Slider(self.slider_ax,
"Time",
0,
self.n,
valinit=ss_plotdata.n,
valstep=1,
valfmt=("%0.0f " + ss_plotdata.lag_units))
self.timestep = 0
self.time_slider.on_changed(self.update_time)
def draw(self, offset):
self.ss_plotdata.update_data(self.entry_ax, self.exit_ax, offset)
self.set_limit()
def update_time(self, val):
self.timestep = int(self.time_slider.val)
self.draw(self.timestep)
self.fig.canvas.draw_idle()
def set_limit(self):
set_all_limits(self.exit_ax, self.min_limit, self.max_limit)
set_all_limits(self.entry_ax, self.min_limit, self.max_limit)
def animate(self, k):
self.timestep += self.ppf
self.timestep %= self.n
print('\r{0:.2f}%'.format(100 * self.timestep / self.n), end='')
self.time_slider.set_val(self.timestep)
return []
def render_animation_to_file(self, outfile):
print("Rendering....")
self.ani = animation.FuncAnimation(
self.fig,
self.animate,
self.ppf * np.arange(0, (self.n + 1) // self.ppf),
interval=20,
repeat=False,
blit=True)
self.ani.save(outfile, writer="ffmpeg")
def render_image_to_file(self, outfile):
self.draw(self.n)
plt.draw()
self.fig.savefig(outfile)
def show_animation(self):
self.ani = animation.FuncAnimation(
self.fig,
self.animate,
self.ppf * np.arange(0, (self.n + 1) // self.ppf),
interval=20,
repeat=True,
blit=True)
plt.show()
def render_image(self, offset):
pass
class PlotFrame(wx.Frame):
"""
PlotFrame is a custom wxPython frame to hold the panel with a
Figure and WxAgg backend canvas for matplotlib plots or other
figures. In this frame:
self is an instance of a wxFrame;
axes is an instance of MPL Axes;
fig is an instance of MPL Figure;
panel is an instance of wxPanel, used for the main panel, to hold
canvas, an instance of MPL FigureCanvasWxAgg.
"""
# Main function to set everything up when the frame is created
def __init__(self, title, pos, size):
"""
This will be executed when an instance of PlotFrame is created.
It is the place to define any globals as "self.<name>".
"""
wx.Frame.__init__(self, None, wx.ID_ANY, title, pos, size)
if len(sys.argv) < 2:
self.filename = ""
else:
self.filename = sys.argv[1]
# set some Boolean flags
self.STOP = False
self.data_loaded = False
self.reverse_play = False
self.step = 1
# Make the main Matplotlib panel for plots
self.create_main_panel() # creates canvas and contents
# Then add wxPython widgets below the MPL canvas
# Layout with box sizers
self.sizer = wx.BoxSizer(wx.VERTICAL)
self.sizer.Add(self.canvas, 1, wx.LEFT | wx.TOP | wx.EXPAND)
self.sizer.AddSpacer(10)
self.sizer.Add(self.toolbar, 0, wx.EXPAND)
self.sizer.AddSpacer(10)
# Make the control panel with a row of buttons
self.create_button_bar()
self.sizer.Add(self.button_bar_sizer, 0, flag=wx.ALIGN_CENTER | wx.TOP)
# Make a Status Bar
self.statusbar = self.CreateStatusBar()
self.sizer.Add(self.statusbar, 0, wx.EXPAND)
self.SetStatusText("Frame created ...")
# -------------------------------------------------------
# set up the Menu Bar
# -------------------------------------------------------
menuBar = wx.MenuBar()
menuFile = wx.Menu() # File menu
menuFile.Append(1, "&Open", "Filename(s) or wildcard list to plot")
menuFile.Append(3, "Save", "Save plot as a PNG image")
menuFile.AppendSeparator()
menuFile.Append(10, "E&xit")
menuBar.Append(menuFile, "&File")
menuHelp = wx.Menu() # Help menu
menuHelp.Append(11, "&About Netview")
menuHelp.Append(12, "&Usage and Help")
menuHelp.Append(13, "Program &Info")
menuBar.Append(menuHelp, "&Help")
self.SetMenuBar(menuBar)
self.panel.SetSizer(self.sizer)
self.sizer.Fit(self)
# -------------------------------------------------------
# Bind the menu items to functions
# -------------------------------------------------------
self.Bind(wx.EVT_MENU, self.OnOpen, id=1)
self.Bind(wx.EVT_MENU, self.OnSave, id=3)
self.Bind(wx.EVT_MENU, self.OnQuit, id=10)
self.Bind(wx.EVT_MENU, self.OnAbout, id=11)
self.Bind(wx.EVT_MENU, self.OnUsage, id=12)
self.Bind(wx.EVT_MENU, self.OnInfo, id=13)
# methods defined below to get and plot the data
# Normally do the plot on request, and not here
# self.get_data_params()
# self.init_plot()
# self.get_xyt_data()
# plot_data()
# ---------- end of __init__ ----------------------------
# -------------------------------------------------------
# Function to make the main Matplotlib panel for plots
# -------------------------------------------------------
def create_main_panel(self):
""" create_main_panel creates the main mpl panel with instances of:
* mpl Canvas
* mpl Figure
* mpl Figure
* mpl Axes with subplot
* mpl Widget class Sliders and Button
* mpl navigation toolbar
self.axes is the instance of MPL Axes, and is where it all happens
"""
self.panel = wx.Panel(self)
# Create the mpl Figure and FigCanvas objects.
# 3.5 x 5 inches, 100 dots-per-inch
#
self.dpi = 100
self.fig = Figure((3.5, 5.0), dpi=self.dpi)
self.canvas = FigCanvas(self.panel, wx.ID_ANY, self.fig)
# Since we have only one plot, we could use add_axes
# instead of add_subplot, but then the subplot
# configuration tool in the navigation toolbar wouldn't work.
self.axes = self.fig.add_subplot(111)
# (111) == (1,1,1) --> row 1, col 1, Figure 1)
# self.axes.set_title("View from: "+self.filename)
# Now create some sliders below the plot after making room
self.fig.subplots_adjust(left=0.1, bottom=0.20)
self.axtmin = self.fig.add_axes([0.2, 0.10, 0.5, 0.03])
self.axtmax = self.fig.add_axes([0.2, 0.05, 0.5, 0.03])
self.stmin = Slider(self.axtmin, 't_min:', 0.0, 1.0, valinit=0.0)
self.stmax = Slider(self.axtmax, 't_max:', 0.0, 1.0, valinit=1.0)
self.stmin.on_changed(self.update_trange)
self.stmax.on_changed(self.update_trange)
self.axbutton = self.fig.add_axes([0.8, 0.07, 0.1, 0.07])
self.reset_button = Button(self.axbutton, 'Reset')
self.reset_button.color = 'skyblue'
self.reset_button.hovercolor = 'lightblue'
self.reset_button.on_clicked(self.reset_trange)
# Create the navigation toolbar, tied to the canvas
self.toolbar = NavigationToolbar(self.canvas)
def update_trange(self, event):
self.t_min = self.stmin.val
self.t_max = self.stmax.val
# print(self.t_min, self.t_max)
def reset_trange(self, event):
self.stmin.reset()
self.stmax.reset()
def create_button_bar(self):
"""
create_button_bar makes a control panel bar with buttons and
toggles for
New Data - Play - STOP - Single Step - Forward/Back - Normal/Fast
It does not create a Panel container, but simply creates Button
objects with bindings, and adds them to a horizontal BoxSizer
self.button_bar_sizer. This is added to the PlotFrame vertical
BoxSizer, after the MPL canvas, during initialization of the frame.
"""
rewind_button = wx.Button(self.panel, -1, "New Data")
self.Bind(wx.EVT_BUTTON, self.OnRewind, rewind_button)
replot_button = wx.Button(self.panel, -1, "Play")
self.Bind(wx.EVT_BUTTON, self.OnReplot, replot_button)
sstep_button = wx.Button(self.panel, -1, "Single Step")
self.Bind(wx.EVT_BUTTON, self.OnSstep, sstep_button)
stop_button = wx.Button(self.panel, -1, "STOP")
self.Bind(wx.EVT_BUTTON, self.OnStop, stop_button)
# The toggle buttons need to be globally accessible
self.forward_toggle = wx.ToggleButton(self.panel, -1, "Forward")
self.forward_toggle.SetValue(True)
self.forward_toggle.SetLabel("Forward")
self.Bind(wx.EVT_TOGGLEBUTTON, self.OnForward, self.forward_toggle)
self.fast_toggle = wx.ToggleButton(self.panel, -1, " Normal ")
self.fast_toggle.SetValue(True)
self.fast_toggle.SetLabel(" Normal ")
self.Bind(wx.EVT_TOGGLEBUTTON, self.OnFast, self.fast_toggle)
# Set button colors to some simple colors that are likely
# to be independent on X11 color definitions. Some nice
# bit maps (from a media player skin?) should be used
# or the buttons and toggle state colors in OnFast() below
rewind_button.SetBackgroundColour('skyblue')
replot_button.SetBackgroundColour('skyblue')
sstep_button.SetBackgroundColour('skyblue')
stop_button.SetBackgroundColour('skyblue')
self.forward_toggle.SetForegroundColour('black')
self.forward_toggle.SetBackgroundColour('yellow')
self.fast_toggle.SetForegroundColour('black')
self.fast_toggle.SetBackgroundColour('yellow')
self.button_bar_sizer = wx.BoxSizer(wx.HORIZONTAL)
flags = wx.ALIGN_CENTER | wx.ALL
self.button_bar_sizer.Add(rewind_button, 0, border=3, flag=flags)
self.button_bar_sizer.Add(replot_button, 0, border=3, flag=flags)
self.button_bar_sizer.Add(sstep_button, 0, border=3, flag=flags)
self.button_bar_sizer.Add(stop_button, 0, border=3, flag=flags)
self.button_bar_sizer.Add(self.forward_toggle, 0, border=3, flag=flags)
self.button_bar_sizer.Add(self.fast_toggle, 0, border=3, flag=flags)
# -------------------------------------------------------
# Functions to generate or read (x,y) data and plot it
# -------------------------------------------------------
def get_data_params(self):
# These parameters would normally be provided in a file header,
# past as arguments in a function, or from other file information
# Next version will bring up a dialog for dt NX NY if no file header
# Here check to see if a filename should be entered from File/Open
# self.filename = 'Ex_net_Vm_0001.txt'
if len(self.filename) == 0:
# fake a button press of File/Open
self.OnOpen(wx.EVT_BUTTON)
# should check here if file exists as specified [path]/filename
# assume it is a bzip2 compressed file
try:
fp = bz2.BZ2File(self.filename)
line = fp.readline()
except IOError:
# then assume plain text
fp = open(self.filename)
line = fp.readline()
fp.close()
# check if first line is a header line starting with '#'
header = line.split()
if header[0][0] == "#":
self.Ntimes = int(header[1])
self.t_min = float(header[2])
self.dt = float(header[3])
self.NX = int(header[4])
self.NY = int(header[5])
else:
pdentry = self.ParamEntryDialog()
if pdentry.ShowModal() == wx.ID_OK:
self.Ntimes = int(pdentry.Ntimes_dialog.entry.GetValue())
self.t_min = float(pdentry.tmin_dialog.entry.GetValue())
self.dt = float(pdentry.dt_dialog.entry.GetValue())
self.NX = int(pdentry.NX_dialog.entry.GetValue())
self.NY = int(pdentry.NY_dialog.entry.GetValue())
print 'Ntimes = ', self.Ntimes, ' t_min = ', self.t_min
print 'NX = ', self.NX, ' NY = ', self.NY
pdentry.Destroy()
self.t_max = (self.Ntimes - 1) * self.dt
# reset slider max and min
self.stmin.valmax = self.t_max
self.stmin.valinit = self.t_min
self.stmax.valmax = self.t_max
self.stmax.valinit = self.t_max
self.stmax.set_val(self.t_max)
self.stmin.reset()
self.stmax.reset()
fp.close()
def init_plot(self):
'''
init_plot creates the initial plot display. A normal MPL plot
would be created here with a command "self.axes.plot(x, y)" in
order to create a plot of points in the x and y arrays on the
Axes subplot. Here, we create an AxesImage instance with
imshow(), instead. The initial image is a blank one of the
proper dimensions, filled with zeroes.
'''
self.t_max = (self.Ntimes - 1) * self.dt
self.axes.set_title("View of " + self.filename)
# Note that NumPy array (row, col) = image (y, x)
data0 = np.zeros((self.NY, self.NX))
# Define a 'cold' to 'hot' color scale based in GENESIS 2 'hot'
hotcolors = [
'#000032', '#00003c', '#000046', '#000050', '#00005a', '#000064',
'#00006e', '#000078', '#000082', '#00008c', '#000096', '#0000a0',
'#0000aa', '#0000b4', '#0000be', '#0000c8', '#0000d2', '#0000dc',
'#0000e6', '#0000f0', '#0000fa', '#0000ff', '#000af6', '#0014ec',
'#001ee2', '#0028d8', '#0032ce', '#003cc4', '#0046ba', '#0050b0',
'#005aa6', '#00649c', '#006e92', '#007888', '#00827e', '#008c74',
'#00966a', '#00a060', '#00aa56', '#00b44c', '#00be42', '#00c838',
'#00d22e', '#00dc24', '#00e61a', '#00f010', '#00fa06', '#00ff00',
'#0af600', '#14ec00', '#1ee200', '#28d800', '#32ce00', '#3cc400',
'#46ba00', '#50b000', '#5aa600', '#649c00', '#6e9200', '#788800',
'#827e00', '#8c7400', '#966a00', '#a06000', '#aa5600', '#b44c00',
'#be4200', '#c83800', '#d22e00', '#dc2400', '#e61a00', '#f01000',
'#fa0600', '#ff0000', '#ff0a00', '#ff1400', '#ff1e00', '#ff2800',
'#ff3200', '#ff3c00', '#ff4600', '#ff5000', '#ff5a00', '#ff6400',
'#ff6e00', '#ff7800', '#ff8200', '#ff8c00', '#ff9600', '#ffa000',
'#ffaa00', '#ffb400', '#ffbe00', '#ffc800', '#ffd200', '#ffdc00',
'#ffe600', '#fff000', '#fffa00', '#ffff00', '#ffff0a', '#ffff14',
'#ffff1e', '#ffff28', '#ffff32', '#ffff3c', '#ffff46', '#ffff50',
'#ffff5a', '#ffff64', '#ffff6e', '#ffff78', '#ffff82', '#ffff8c',
'#ffff96', '#ffffa0', '#ffffaa', '#ffffb4', '#ffffbe', '#ffffc8',
'#ffffd2', '#ffffdc', '#ffffe6', '#fffff0'
]
cmap = matplotlib.colors.ListedColormap(hotcolors)
self.im = self.axes.imshow(data0, cmap=cmap, origin='lower')
# http://matplotlib.sourceforge.net/examples/pylab_examples/show_colormaps.html
# shows examples to use as a 'cold' to 'hot' mapping of value to color
# cm.jet, cm.gnuplot and cm.afmhot are good choices, but are unlike G2 'hot'
self.im.cmap = cmap
# Not sure how to properly add a colorbar
# self.cb = self.fig.colorbar(self.im, orientation='vertical')
# refresh the canvas
self.canvas.draw()
def get_xyt_data(self):
# Create scaled (0-1) luminance(x,y) array from ascii G-2 disk_out file
# get the data to plot from the specified filename
# Note that NumPy loadtxt transparently deals with bz2 compression
self.SetStatusText('Data loading - please wait ....')
rawdata = np.loadtxt(self.filename)
# Note the difference between NumPy [row, col] order and network
# x-y grid (x, y) = (col, row). We want a NumPy NY x NX, not
# NX x NY, array to be used by the AxesImage object.
xydata = np.resize(rawdata, (self.Ntimes, self.NY, self.NX))
# imshow expects the data to be scaled to range 0-1.
Vmin = xydata.min()
Vmax = xydata.max()
self.ldata = (xydata - Vmin) / (Vmax - Vmin)
self.data_loaded = True
self.SetStatusText('Data has been loaded - click Play')
def plot_data(self):
''' plot_data() shows successive frames of the data that was loaded
into the ldata array. Creating a new self.im AxesImage instance
for each frame is extremely slow, so the set_data method of
AxesImage is used to load new data into the existing self.im for
each frame. Normally 'self.canvas.draw()' would be used to
display a frame, but redrawing the entire canvas, redraws the
axes, labels, sliders, buttons, or anything else on the canvas.
This uses a method taken from an example in Ch 7, p. 192
Matplotlib for Python developers, with draw_artist() and blit()
redraw only the part that was changed.
'''
if self.data_loaded == False:
# bring up a warning dialog
msg = """
Data for plotting has not been loaded!
Please enter the file to plot with File/Open, unless
it was already specified, and then click on 'New Data'
to load the data to play back, before clicking 'Play'.
"""
wx.MessageBox(msg, "Plot Warning", wx.OK | wx.ICON_ERROR, self)
return
# set color limits
self.im.set_clim(0.0, 1.0)
self.im.set_interpolation('nearest')
# 'None' is is slightly faster, but not implemented for MPL ver < 1.1
# self.im.set_interpolation('None')
# do an initial draw, then save the empty figure axes
self.canvas.draw()
# self.bg = self.canvas.copy_from_bbox(self.axes.bbox)
# However the save and restore is only needed if I change
# axes legends, etc. The draw_artist(artist), and blit
# are much faster than canvas.draw() and are sufficient.
print 'system time (seconds) = ', time.time()
# round frame_min down and frame_max up for the time window
frame_min = int(self.t_min / self.dt)
frame_max = min(int(self.t_max / self.dt) + 1, self.Ntimes)
frame_step = self.step
# Displaying simulation time to the status bar is much faster
# than updating a slider progress bar, but location isn't optimum.
# The check for the STOP button doesn't work because the button
# click is not registered until this function exits.
# check to see if self.reverse_play == True
# then interchange frame_min, frame_max, and use negative step
if self.reverse_play == True:
frame_min = min(int(self.t_max / self.dt) + 1, self.Ntimes) - 1
frame_max = int(self.t_min / self.dt) - 1
frame_step = -self.step
for frame_num in range(frame_min, frame_max, frame_step):
self.SetStatusText('time: ' + str(frame_num * self.dt))
if self.STOP == True:
self.t_min = frame_num * self.dt
# set t_min slider ?
self.STOP = False
break
self.im.set_data(self.ldata[frame_num])
self.axes.draw_artist(self.im)
self.canvas.blit(self.axes.bbox)
print 'system time (seconds) = ', time.time()
# ------------------------------------------------------------------
# Define the classes and functions for getting parameter values
# --------------------------------------------------------------
class ParamEntryDialog(wx.Dialog):
def __init__(self):
wx.Dialog.__init__(self, None, wx.ID_ANY)
self.SetSize((250, 200))
self.SetTitle('Enter Data File Parameters')
vbox = wx.BoxSizer(wx.VERTICAL)
self.Ntimes_dialog = XDialog(self)
self.Ntimes_dialog.entry_label.SetLabel('Number of entries')
self.Ntimes_dialog.entry.ChangeValue(str(2501))
self.tmin_dialog = XDialog(self)
self.tmin_dialog.entry_label.SetLabel('Start time (sec)')
self.tmin_dialog.entry.ChangeValue(str(0.0))
self.dt_dialog = XDialog(self)
self.dt_dialog.entry_label.SetLabel('Output time step (sec)')
self.dt_dialog.entry.ChangeValue(str(0.0002))
self.NX_dialog = XDialog(self)
self.NX_dialog.entry_label.SetLabel('Number of cells on x-axis')
self.NX_dialog.entry.ChangeValue(str(32))
self.NY_dialog = XDialog(self)
self.NY_dialog.entry_label.SetLabel('Number of cells on y-axis')
self.NY_dialog.entry.ChangeValue(str(32))
vbox.Add(self.Ntimes_dialog, 0, wx.EXPAND | wx.ALL, border=5)
vbox.Add(self.tmin_dialog, 0, wx.EXPAND | wx.ALL, border=5)
vbox.Add(self.dt_dialog, 0, wx.EXPAND | wx.ALL, border=5)
vbox.Add(self.NX_dialog, 0, wx.EXPAND | wx.ALL, border=5)
vbox.Add(self.NY_dialog, 0, wx.EXPAND | wx.ALL, border=5)
okButton = wx.Button(self, wx.ID_OK, 'Ok')
# vbox.Add(okButton,flag=wx.ALIGN_CENTER|wx.TOP|wx.BOTTOM, border=10)
vbox.Add(okButton, flag=wx.ALIGN_CENTER, border=10)
self.SetSizer(vbox)
self.SetSizerAndFit(vbox)
# ------------------------------------------------------------------
# Define the functions executed on menu choices
# ---------------------------------------------------------------
def OnQuit(self, event):
self.Close()
def OnSave(self, event):
file_choices = "PNG (*.png)|*.png"
dlg = wx.FileDialog(self,
message="Save plot as...",
defaultDir=os.getcwd(),
defaultFile="plot.png",
wildcard=file_choices,
style=wx.SAVE)
if dlg.ShowModal() == wx.ID_OK:
path = dlg.GetPath()
self.canvas.print_figure(path, dpi=self.dpi)
# self.flash_status_message("Saved to %s" % path)
def OnAbout(self, event):
msg = """
G-3 Netview ver. 1.7
Netview is a stand-alone Python application for viewing
the output of GENESIS 2 and 3 network simulations.
It is intended to replace GENESIS 2 SLI scripts that use the
XODUS 'xview' widget.
The design and operation is based on the G3Plot application
for creating 2D plots of y(t) or y(x) from data files.
Unlike G3Plot, the image created with Netview is an animated
representation of a rectangular network with colored squares
used to indicate the value of some variable at that position
and time. Typically, this would be the membrane potenial of
a cell soma, or a synaptic current in a dendrite segment.
Help/Usage gives HTML help for using Netview.
This is the main Help page.
Help/Program Info provides some information about the
objects and functions, and the wxPython and matplotlib
classes used here.
Dave Beeman, August 2012
"""
dlg = wx.MessageDialog(self, msg, "About G-3 Netview",
wx.OK | wx.ICON_QUESTION)
dlg.ShowModal()
dlg.Destroy()
def OnOpen(self, event):
dlg = wx.TextEntryDialog(self,
"File with x,y data to plot",
"File Open",
self.filename,
style=wx.OK | wx.CANCEL)
if dlg.ShowModal() == wx.ID_OK:
self.filename = dlg.GetValue()
# A new filename has been entered, but the data has not been read
self.data_loaded = False
# print "You entered: %s" % self.filename
dlg.Destroy()
# This starts with the long string of HTML to display
class UsageFrame(wx.Frame):
text = """
<HTML>
<HEAD></HEAD>
<BODY BGCOLOR="#D6E7F7">
<CENTER><H1>Using G-3 Netview</H1></CENTER>
<H2>Introduction and Quick Start</H2>
<p>Netview is a stand-alone Python application for viewing the output of
GENESIS 2 and 3 network simulations. It is intended to replace GENESIS 2
SLI scripts that use the XODUS 'xview' widget.</p>
<p>The design and operation is based on the G3Plot application for creating 2D
plots of y(t) or y(x) from data files. As with G3Plot, the main class
PlotFrame uses a basic wxPython frame to embed a matplotlib figure for
plotting. It defines some basic menu items and a control panel of buttons
and toggles, each with bindings to a function to execute on a mouse click.</p>
<p>Unlike G3Plot, the image created with Netview is an animated
representation of a rectangular network with colored squares
used to indicate the value of some variable at that position
and time. Typically, this would be the membrane potenial of
a cell soma, or a synaptic current in a dendrite segment.</p>
<h2>Usage</h2>
<p>The Menu Bar has <em>File/Open</em>, <em>File/Save</em>, and
<em>File/Exit</em> choices. The Help Menu choices <em>About</em> and
<em>Usage</em> give further information. The <em>Program Info</em>
selection shows code documentation that is contained in some of the main
function <em>docstrings</em>.</p>
<p>After starting the <em>netview</em> program, enter a data file name
in the dialog for File/Open, unless the filename was given as a
command line argument. Then click on <strong>New Data</strong> to load the new
data and initialize the plot. When the plot is cleared to black,
press <strong>Play</strong>.</p>
<p>The file types recognized are plain text or text files compressed with
bzip2. The expected data format is one line for each output time step,
with each line having the membrane potential value of each cell in the net.
No time value should be given on the line. In order to properly display
the data, netview needs some additional information about the network and
the data. This can optionally be contained in a header line that precedes
the data. If a header is not detected, a dialog will appear asking for the
needed parameters.</p>
<p>It is assumed that the cells are arranged on a NX x NY grid, numbered
from 0 (bottom left corner) to NX*NY - 1 (upper right corner).
In order to provide this information to netview, the data file should
begin with a header line of the form:</p>
<pre>
#optional_RUNID_string Ntimes start_time dt NX NY SEP_X SEP_Y x0 y0 z0
</pre>
<p>The line must start with "#" and can optionally be followed immediately by any
string. Typically this is some identification string generated by the
simulation run. The following parameters, separated by blanks or any
whitespace, are:</p>
<ul>
<li>Ntimes - the number of lines in the file, exclusive of the header</li>
<li>start_time - the simulation time for the first data line (default 0.0)</li>
<li>dt - the time step used for output</li>
<li>NX, NY - the integer dimensions of the network</li>
<li>SEP_X, SEP_Y - the x,y distances between cells (optional)</li>
<li>x0, y0, z0 - the location of the compartment (data source) relative to the
cell origin</li>
</ul>
<p>The RUNID string and the last five parameters are not read or used
by netview. These are available for other data analysis tools that
need a RUNID and the location of each source.</p>
<p>The slider bars can be used to set a time window for display, and the
<strong>Reset</strong> button can set t_min and t_max back to the defaults.
Use the <strong>Forward/Back</strong> toggle to reverse direction of
<strong>Play</strong>, and the <strong>Normal/Fast</strong> toggle to show
every tenth frame.</p> <p>The <strong>Single Step</strong> button can be
used to advance a single step at a time (or 10, if in 'Fast' mode).</p>
<p>The <strong>STOP</strong> button is currently not implemented</p>
<p>To plot different data, enter a new filename with <strong>File/Open</strong> and
repeat with <strong>New Data</strong> and <strong>Play</strong>.</p>
<HR>
</BODY>
</HTML>
"""
def __init__(self, parent):
wx.Frame.__init__(self,
parent,
-1,
"Usage and Help",
size=(640, 600),
pos=(400, 100))
html = wx.html.HtmlWindow(self)
html.SetPage(self.text)
panel = wx.Panel(self, -1)
button = wx.Button(panel, wx.ID_OK, "Close")
self.Bind(wx.EVT_BUTTON, self.OnCloseMe, button)
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add(html, 1, wx.EXPAND | wx.ALL, 5)
sizer.Add(panel, 0, wx.ALIGN_CENTER | wx.ALL, 5)
self.SetSizer(sizer)
self.Layout()
def OnCloseMe(self, event):
self.Close(True)
# ----------- end of class UsageFrame ---------------
def OnUsage(self, event):
usagewin = self.UsageFrame(self)
usagewin.Show(True)
def OnInfo(self, event):
msg = "Program information for PlotFrame obtained from docstrings:"
msg += "\n" + self.__doc__ + "\n" + self.create_main_panel.__doc__
msg += self.create_button_bar.__doc__
msg += self.init_plot.__doc__
msg += self.plot_data.__doc__
dlg = wx.lib.dialogs.ScrolledMessageDialog(self, msg,
"PlotFrame Documentation")
dlg.ShowModal()
# ---------------------------------------------------------------
# Define the functions executed on control button click
# ---------------------------------------------------------------
def OnRewind(self, event):
self.get_data_params()
self.init_plot()
self.get_xyt_data()
def OnReplot(self, event):
self.plot_data()
self.canvas.draw()
def OnSstep(self, event):
if self.data_loaded == False:
# bring up a warning dialog
msg = """
Data for plotting has not been loaded!
Please enter the file to plot with File/Open, unless
it was already specified, and then click on 'New Data'
to load the data to play back, before clicking 'Play'.
"""
wx.MessageBox(msg, "Plot Warning", wx.OK | wx.ICON_ERROR, self)
return
self.t_max = min(self.t_max + self.dt, (self.Ntimes - 1) * self.dt)
self.stmax.set_val(self.t_max)
frame_num = int(self.t_max / self.dt)
self.SetStatusText('time: ' + str(frame_num * self.dt))
self.im.set_data(self.ldata[frame_num])
self.axes.draw_artist(self.im)
self.canvas.blit(self.axes.bbox)
def OnStop(self, event):
self.STOP = 'True'
def OnForward(self, event):
state = self.forward_toggle.GetValue()
if state:
self.reverse_play = False
self.forward_toggle.SetLabel("Forward ")
self.forward_toggle.SetForegroundColour('black')
self.forward_toggle.SetBackgroundColour('yellow')
else:
self.reverse_play = True
self.forward_toggle.SetLabel(" Back ")
self.forward_toggle.SetForegroundColour('red')
self.forward_toggle.SetBackgroundColour('green')
def OnFast(self, event):
state = self.fast_toggle.GetValue()
if state:
# print state
self.fast_toggle.SetLabel(" Normal ")
self.fast_toggle.SetForegroundColour('black')
self.fast_toggle.SetBackgroundColour('yellow')
self.step = 1
else:
# print state
self.fast_toggle.SetLabel(" Fast ")
self.fast_toggle.SetForegroundColour('red')
self.fast_toggle.SetBackgroundColour('green')
self.step = 10
class ParametricEQSelector:
def __init__(self):
self._fs = 48000
self._nbands = 7
self._params = []
for i in range(0, self._nbands):
self._params.append({'center': 0, 'resonance': 1.0/math.sqrt(2.0), 'dbgain': 0})
self._selected_band = 0
self._blocksize = 512
self._nfft = int(self._blocksize / 2)
self._impulse_response = [0] * self._blocksize
self._freq_response_real = [0] * self._blocksize
self._freq_response_imag = [0] * self._blocksize
self._response = [0] * self._blocksize
self._plot_db = True
self._eq = yodel.filter.ParametricEQ(self._fs, self._nbands)
self._create_plot()
self._create_plot_controls()
self.select_band('Band ' + str(self._selected_band+1))
def _create_plot(self):
self._fig, self._ax = plt.subplots()
self._ax.set_title('Parametric Equalizer Design')
self._ax.grid()
plt.subplots_adjust(bottom=0.3)
self._update_filter_response()
self._x_axis = [i*(self._fs/2/self._nfft) for i in range(0, self._nfft)]
self._y_axis = self._response[0:self._nfft]
self._l_center, = self._ax.plot(self._x_axis, [0] * self._nfft, 'k')
self._l_fr, = self._ax.plot(self._x_axis, self._y_axis, 'b')
self._rescale_plot()
def _create_plot_controls(self):
self._dbrax = plt.axes([0.12, 0.05, 0.13, 0.10])
self._dbradio = RadioButtons(self._dbrax, ('Amplitude', 'Phase'))
self._dbradio.on_clicked(self.set_plot_style)
self._rax = plt.axes([0.27, 0.03, 0.15, 0.20])
bands_list = []
for i in range(1, self._nbands+1):
bands_list.append('Band ' + str(i))
self._radio = RadioButtons(self._rax, tuple(bands_list))
self._radio.on_clicked(self.select_band)
self._sfax = plt.axes([0.6, 0.19, 0.2, 0.03])
self._sqax = plt.axes([0.6, 0.12, 0.2, 0.03])
self._sdbax = plt.axes([0.6, 0.05, 0.2, 0.03])
self._fcslider = Slider(self._sfax, 'Cut-off frequency', 0, self._fs/2, valinit = self._params[self._selected_band]['center'])
self._qslider = Slider(self._sqax, 'Q factor', 0.01, 10.0, valinit = self._params[self._selected_band]['resonance'])
self._dbslider = Slider(self._sdbax, 'dB gain', -20.0, 20.0, valinit = self._params[self._selected_band]['dbgain'])
self._fcslider.on_changed(self.set_center_frequency)
self._qslider.on_changed(self.set_resonance)
self._dbslider.on_changed(self.set_dbgain)
def _rescale_plot(self):
if self._plot_db:
self._ax.set_ylim(-30, 30)
else:
self._ax.set_ylim(- 200, 200)
plt.draw()
def _plot_frequency_response(self, redraw=True):
self._update_filter_response()
self._y_axis = self._response[0:self._nfft]
self._l_fr.set_ydata(self._y_axis)
if redraw:
plt.draw()
def _plot_range_limits(self, redraw=True):
self._l_center.set_ydata([0] * self._nfft)
if redraw:
plt.draw()
def set_plot_style(self, style):
if style == 'Phase':
self._plot_db = False
elif style == 'Amplitude':
self._plot_db = True
self._plot_range_limits(False)
self._plot_frequency_response(False)
self._rescale_plot()
def select_band(self, band):
idx = band.split(' ')
self._selected_band = int(idx[1]) - 1
self._fcslider.set_val(self._params[self._selected_band]['center'])
self._qslider.set_val(self._params[self._selected_band]['resonance'])
self._dbslider.set_val(self._params[self._selected_band]['dbgain'])
def set_center_frequency(self, val):
self._params[self._selected_band]['center'] = val
self._set_band(self._selected_band)
self._plot_frequency_response()
def set_resonance(self, val):
self._params[self._selected_band]['resonance'] = val
self._set_band(self._selected_band)
self._plot_frequency_response()
def set_dbgain(self, val):
self._params[self._selected_band]['dbgain'] = val
self._set_band(self._selected_band)
self._plot_frequency_response()
def _set_band(self, band):
center = self._params[band]['center']
resonance = self._params[band]['resonance']
dbgain = self._params[band]['dbgain']
self._eq.set_band(band, center, resonance, dbgain)
def _update_filter_response(self):
self._impulse_response = impulse_response(self._eq, self._blocksize)
self._freq_response_real, self._freq_response_imag = frequency_response(self._impulse_response)
if self._plot_db:
self._response = amplitude_response(self._freq_response_real, self._freq_response_imag)
else:
self._response = phase_response(self._freq_response_real, self._freq_response_imag)
class EnergyPlusModel(metaclass=ABCMeta):
def __init__(self, model_file, log_dir=None, verbose=False):
self.log_dir = log_dir
self.model_basename = os.path.splitext(os.path.basename(model_file))[0]
self.setup_spaces()
self.action = 0.5 * (self.action_space.low + self.action_space.high)
self.action_prev = self.action
self.raw_state = None
self.verbose = verbose
self.timestamp_csv = None
self.sl_episode = None
# Progress data
self.num_episodes = 0
self.num_episodes_last = 0
self.reward = None
self.reward_mean = None
def reset(self):
pass
# Parse date/time format from EnergyPlus and return datetime object with correction for 24:00 case
def _parse_datetime(self, dstr):
# ' MM/DD HH:MM:SS' or 'MM/DD HH:MM:SS'
# Dirty hack
if dstr[0] != ' ':
dstr = ' ' + dstr
# year = 2017
year = 2013 # for CHICAGO_IL_USA TMY2-94846
month = int(dstr[1:3])
day = int(dstr[4:6])
hour = int(dstr[8:10])
minute = int(dstr[11:13])
sec = 0
msec = 0
if hour == 24:
hour = 0
dt = datetime(year, month, day, hour, minute, sec,
msec) + timedelta(days=1)
else:
dt = datetime(year, month, day, hour, minute, sec, msec)
return dt
# Convert list of date/time string to list of datetime objects
def _convert_datetime24(self, dates):
# ' MM/DD HH:MM:SS'
dates_new = []
for d in dates:
# year = 2017
# month = int(d[1:3])
# day = int(d[4:6])
# hour = int(d[8:10])
# minute = int(d[11:13])
# sec = 0
# msec = 0
# if hour == 24:
# hour = 0
# d_new = datetime(year, month, day, hour, minute, sec, msec) + dt.timedelta(days=1)
# else:
# d_new = datetime(year, month, day, hour, minute, sec, msec)
# dates_new.append(d_new)
dates_new.append(self._parse_datetime(d))
return dates_new
# Generate x_pos and x_labels
def generate_x_pos_x_labels(self, dates):
time_delta = self._parse_datetime(dates[1]) - self._parse_datetime(
dates[0])
x_pos = []
x_labels = []
for i, d in enumerate(dates):
dt = self._parse_datetime(d) - time_delta
if dt.hour == 0 and dt.minute == 0:
x_pos.append(i)
x_labels.append(dt.strftime('%m/%d'))
return x_pos, x_labels
def set_action(self, action):
# In TPRO/POP1/POP2 in baseline, action seems to be normalized to [-1.0, 1.0].
# So it must be scaled back into action_space by the environment.
assert action.shape == self.action_space.low.shape, 'Invalid action {}'.format(
action)
self.action_prev = self.action
self.action = action
self.action = np.clip(self.action, self.action_space.low,
self.action_space.high)
# self.action_prev = self.action
# self.action = self.action_space.low + (normalized_action + 1.) * 0.5 * (
# self.action_space.high - self.action_space.low)
# self.action = np.clip(self.action, self.action_space.low, self.action_space.high)
@abstractmethod
def setup_spaces(self):
pass
# Need to handle the case that raw_state is None
@abstractmethod
def set_raw_state(self, raw_state):
pass
def get_state(self):
return self.format_state(self.raw_state)
@abstractmethod
def compute_reward(self):
pass
@abstractmethod
def format_state(self, raw_state):
pass
# --------------------------------------------------
# Plotting staffs follow
# --------------------------------------------------
def plot(self, log_dir='', csv_file='', **kwargs):
if log_dir is not '':
if not os.path.isdir(log_dir):
print('energyplus_model.plot: {} is not a directory'.format(
log_dir))
return
print('energyplus_plot.plot log={}'.format(log_dir))
self.log_dir = log_dir
self.show_progress()
else:
if not os.path.isfile(csv_file):
print(
'energyplus_model.plot: {} is not a file'.format(csv_file))
return
print('energyplus_model.plot csv={}'.format(csv_file))
self.read_episode(csv_file)
plt.rcdefaults()
plt.rcParams['font.size'] = 6
plt.rcParams['lines.linewidth'] = 1.0
plt.rcParams['legend.loc'] = 'lower right'
self.fig = plt.figure(1, figsize=(16, 10))
self.plot_episode(csv_file)
plt.show()
# Show convergence
def show_progress(self):
self.monitor_file = self.log_dir + '/monitor.csv'
# Read progress file
if not self.read_monitor_file():
print('Progress data is missing')
sys.exit(1)
# Initialize graph
plt.rcdefaults()
plt.rcParams['font.size'] = 6
plt.rcParams['lines.linewidth'] = 1.0
plt.rcParams['legend.loc'] = 'lower right'
self.fig = plt.figure(1, figsize=(16, 10))
# Show widgets
axcolor = 'lightgoldenrodyellow'
self.axprogress = self.fig.add_axes([0.15, 0.10, 0.70, 0.15],
facecolor=axcolor)
self.axslider = self.fig.add_axes([0.15, 0.04, 0.70, 0.02],
facecolor=axcolor)
axfirst = self.fig.add_axes([0.15, 0.01, 0.03, 0.02])
axlast = self.fig.add_axes([0.82, 0.01, 0.03, 0.02])
axprev = self.fig.add_axes([0.46, 0.01, 0.03, 0.02])
axnext = self.fig.add_axes([0.51, 0.01, 0.03, 0.02])
# Slider is drawn in plot_progress()
# First/Last button
self.button_first = Button(axfirst,
'First',
color=axcolor,
hovercolor='0.975')
self.button_first.on_clicked(self.first_episode_num)
self.button_last = Button(axlast,
'Last',
color=axcolor,
hovercolor='0.975')
self.button_last.on_clicked(self.last_episode_num)
# Next/Prev button
self.button_prev = Button(axprev,
'Prev',
color=axcolor,
hovercolor='0.975')
self.button_prev.on_clicked(self.prev_episode_num)
self.button_next = Button(axnext,
'Next',
color=axcolor,
hovercolor='0.975')
self.button_next.on_clicked(self.next_episode_num)
# Timer
self.timer = self.fig.canvas.new_timer(interval=1000)
self.timer.add_callback(self.check_update)
self.timer.start()
# Progress data
self.axprogress.set_xmargin(0)
self.axprogress.set_xlabel('Episodes')
self.axprogress.set_ylabel('Reward')
self.axprogress.grid(True)
self.plot_progress()
# Plot latest episode
self.update_episode(self.num_episodes - 1)
plt.show()
def check_update(self):
if self.read_monitor_file():
self.plot_progress()
def plot_progress(self):
# Redraw all lines
self.axprogress.lines = []
self.axprogress.plot(self.reward, color='#1f77b4', label='Reward')
# self.axprogress.plot(self.reward_mean, color='#ff7f0e', label='Reward (average)')
self.axprogress.legend()
# Redraw slider
if self.sl_episode is None or int(round(
self.sl_episode.val)) == self.num_episodes - 2:
cur_ep = self.num_episodes - 1
else:
cur_ep = int(round(self.sl_episode.val))
self.axslider.clear()
# self.sl_episode = Slider(self.axslider, 'Episode (0..{})'.format(self.num_episodes - 1), 0, self.num_episodes - 1, valinit=self.num_episodes - 1, valfmt='%6.0f')
self.sl_episode = Slider(self.axslider,
'Episode (0..{})'.format(self.num_episodes -
1),
0,
self.num_episodes - 1,
valinit=cur_ep,
valfmt='%6.0f')
self.sl_episode.on_changed(self.set_episode_num)
def read_monitor_file(self):
# For the very first call, Wait until monitor.csv is created
if self.timestamp_csv is None:
while not os.path.isfile(self.monitor_file):
time.sleep(1)
self.timestamp_csv = os.stat(
self.monitor_file
).st_mtime - 1 # '-1' is a hack to prevent losing the first set of data
num_ep = 0
ts = os.stat(self.monitor_file).st_mtime
if ts > self.timestamp_csv:
# Monitor file is updated.
self.timestamp_csv = ts
f = open(self.monitor_file)
firstline = f.readline()
assert firstline.startswith('#')
metadata = json.loads(firstline[1:])
assert metadata['env_id'] == "EnergyPlus-v0"
assert set(metadata.keys()) == {
'env_id', 't_start'
}, "Incorrect keys in monitor metadata"
df = pd.read_csv(f, index_col=None)
assert set(df.keys()) == {'l', 't',
'r'}, "Incorrect keys in monitor logline"
f.close()
self.reward = []
self.reward_mean = []
self.episode_dirs = []
self.num_episodes = 0
for rew, len, time_ in zip(df['r'], df['l'], df['t']):
self.reward.append(rew / len)
self.reward_mean.append(rew / len)
self.episode_dirs.append(
self.log_dir +
'/output/episode-{:08d}'.format(self.num_episodes))
self.num_episodes += 1
if self.num_episodes > self.num_episodes_last:
self.num_episodes_last = self.num_episodes
return True
else:
return False
def update_episode(self, ep):
self.plot_episode(ep)
def set_episode_num(self, val):
ep = int(round(self.sl_episode.val))
self.update_episode(ep)
def first_episode_num(self, val):
self.sl_episode.set_val(0)
def last_episode_num(self, val):
self.sl_episode.set_val(self.num_episodes - 1)
def prev_episode_num(self, val):
ep = int(round(self.sl_episode.val))
if ep > 0:
ep -= 1
self.sl_episode.set_val(ep)
def next_episode_num(self, val):
ep = int(round(self.sl_episode.val))
if ep < self.num_episodes - 1:
ep += 1
self.sl_episode.set_val(ep)
def show_statistics(self, title, series):
print('{:25} ave={:5,.2f}, min={:5,.2f}, max={:5,.2f}, std={:5,.2f}'.
format(title, np.average(series), np.min(series), np.max(series),
np.std(series)))
def get_statistics(self, series):
return np.average(series), np.min(series), np.max(series), np.std(
series)
def show_distrib(self, title, series):
dist = [0 for i in range(1000)]
for v in series:
idx = int(math.floor(v * 10))
if idx >= 1000:
idx = 999
if idx < 0:
idx = 0
dist[idx] += 1
print(title)
print(
' degree 0.0-0.9 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9'
)
print(
' -------------------------------------------------------------------------'
)
for t in range(170, 280, 10):
print(' {:4.1f}C {:5.1%} '.format(
t / 10.0,
sum(dist[t:(t + 10)]) / len(series)),
end='')
for tt in range(t, t + 10):
print(' {:5.1%}'.format(dist[tt] / len(series)), end='')
print('')
def get_episode_list(self, log_dir='', csv_file=''):
if (log_dir is not '' and csv_file is not '') or (log_dir is ''
and csv_file is ''):
print('Either one of log_dir or csv_file must be specified')
quit()
if log_dir is not '':
if not os.path.isdir(log_dir):
print('energyplus_model.dump: {} is not a directory'.format(
log_dir))
return
print('energyplus_plot.dump: log={}'.format(log_dir))
# self.log_dir = log_dir
# Make a list of all episodes
# Note: Somethimes csv file is missing in the episode directories
# We accept gziped csv file also.
csv_list = glob(log_dir + '/output/episode-????????/eplusout.csv') \
+ glob(log_dir + '/output/episode-????????/eplusout.csv.gz')
self.episode_dirs = list(
set([os.path.dirname(i) for i in csv_list]))
self.episode_dirs.sort()
self.num_episodes = len(self.episode_dirs)
else: # csv_file != ''
self.episode_dirs = [os.path.dirname(csv_file)]
self.num_episodes = len(self.episode_dirs)
# Model dependent methods
@abstractmethod
def read_episode(self, ep):
pass
@abstractmethod
def plot_episode(self, ep):
pass
@abstractmethod
def dump_timesteps(self, log_dir='', csv_file='', **kwargs):
pass
@abstractmethod
def dump_episodes(self, log_dir='', csv_file='', **kwargs):
pass
class Player(FuncAnimation):
"""
The class makes a player with play, stop, and next buttons and a frame slider.
"""
def __init__(self,
fig,
func,
init_func=None,
fargs=None,
save_count=None,
button_color='yellow',
bg_color='red',
dis_start=0,
dis_stop=100,
pos=(0.125, 0.05),
**kwargs):
"""
initialization
:param fig: matplotlib fifure object
:param func: user-defined function which takes a integer (frame number) as an input
:param init_func: user-defined initial function used by the FuncAnimation class
:param fargs: arguments of func, used by FuncAnimation class
:param save_count: save count arg used by FuncAnimation class
:param button_color: string, color of the buttons of the player
:param bg_color: string, hovercolor of the buttons and slider
:param dis_start: int, start frame number
:param dis_stop: int, stop frame number
:param pos: length 2 tuple, position of the buttons
:param kwargs: kwargs for FuncAnimation class
"""
# setting up the index
self.start_ind = dis_start
self.stop_ind = dis_stop
self.dis_length = self.stop_ind - self.start_ind
self.ind = self.start_ind
self.runs = True
self.forwards = True
self.fig = fig
self.fig.set_facecolor('k')
self.button_color = button_color
self.bg_color = bg_color
self.func = func
self.setup(pos)
FuncAnimation.__init__(self,
self.fig,
self.func,
frames=self.play(),
init_func=init_func,
fargs=fargs,
save_count=save_count,
**kwargs)
@property
def ind(self):
return self._ind
@ind.setter
def ind(self, val):
self._ind = val
self._ind -= self.start_ind
self._ind %= (self.dis_length)
self._ind += self.start_ind
def play(self):
"""
play function
"""
while self.runs:
self.ind = self.ind + self.forwards - (not self.forwards)
self._update()
yield self.ind
def start(self):
self.runs = True
self._update()
self.event_source.start()
def stop(self, event=None):
self.runs = False
self._update()
self.event_source.stop()
def forward(self, event=None):
self.forwards = True
self.start()
def backward(self, event=None):
self.forwards = False
self.start()
def oneforward(self, event=None):
self.forwards = True
self.onestep()
def onebackward(self, event=None):
self.forwards = False
self.onestep()
def onestep(self):
if self.forwards:
self.ind += 1
else:
self.ind -= 1
self.func(self.ind)
self._update()
self.fig.canvas.draw_idle()
def _update(self):
self.slider.set_val(self.ind)
def __set_slider(self, val):
val = int(val)
self.ind = val
#self.func(self.ind)
def setup(self, pos):
"""
Setting up the buttons and the slider
:param pos: length 2 tuple, position of the axes for buttons and tuples
:return:
"""
playerax = self.fig.add_axes([pos[0], pos[1], 0.22, 0.04])
divider = mpl_toolkits.axes_grid1.make_axes_locatable(playerax)
bax = divider.append_axes("right", size="80%", pad=0.05)
sax = divider.append_axes("right", size="80%", pad=0.05)
fax = divider.append_axes("right", size="80%", pad=0.05)
ofax = divider.append_axes("right", size="100%", pad=0.05)
sliderax = self.fig.add_axes(
(pos[0], pos[1] - 0.045, 0.5, 0.04),
facecolor=self.bg_color) # 'lemonchiffon')
self.button_oneback = matplotlib.widgets.Button(
playerax,
color=self.button_color,
hovercolor=self.bg_color,
label='$\u29CF$') # , label=r'$\u29CF$')
self.button_back = matplotlib.widgets.Button(
bax,
color=self.button_color,
hovercolor=self.bg_color,
label='$\u25C0$') # r'$\u25C0$')
self.button_stop = matplotlib.widgets.Button(
sax,
color=self.button_color,
hovercolor=self.bg_color,
label='$\u25A0$') # , label=r'$\u25A0$')
self.button_forward = matplotlib.widgets.Button(
fax,
color=self.button_color,
hovercolor=self.bg_color,
label='$\u25B6$') # , label=r'$\u25B6$')
self.button_oneforward = matplotlib.widgets.Button(
ofax,
color=self.button_color,
hovercolor=self.bg_color,
label='$\u29D0$') # , label=r'$\u29D0$')
self.button_oneback.on_clicked(self.onebackward)
self.slider = Slider(sliderax,
label='',
valfmt='%0.0f',
valmin=0,
valmax=self.stop_ind - 1,
valinit=self.ind,
color='black',
fc=self.button_color) # , snap='True')
self.slider.label.set_color(self.button_color)
# self.slider.valtext.set_color(self.button_color)
self.slider.valtext.set_position((0.5, 0.5))
self.slider.set_val(self.ind)
self.button_back.on_clicked(self.backward)
self.button_stop.on_clicked(self.stop)
self.button_forward.on_clicked(self.forward)
self.button_oneforward.on_clicked(self.oneforward)
self.slider.on_changed(self.__set_slider)
class CubeDisplayBase(ImageDisplay):
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def __init__(self, ax, data, coords=None, **kwargs):
'''
Image display for 3D data. Implements frame slider and image scroll.
Optionally also displays apertures if coordinates provided.
subclasses must implement set_frame, get_frame methods
Parameters
----------
ax : Axes object
Axes on which to display
data : array-like
initial display data
coords : optional, np.ndarray
coordinates of apertures to display. This must be an np.ndarray with
shape (k, N, 2) where k is the number of apertures per frame, and N
is the number of frames
kwargs are passed directly to ImageDisplay.
'''
#setup image display
self.autoscale = kwargs.pop('autoscale', 'percentile') #TODO: move up??
ImageDisplay.__init__(self, ax, data, **kwargs)
#self.coords = coords
#setup frame slider
self._frame = 0
self.fsax = self.divider.append_axes('bottom', size=0.2, pad=0.25)
#TODO: elliminated this SHIT Slider class!!!
self.frame_slider = Slider(self.fsax, 'frame', 0, len(self), valfmt='%d')
self.frame_slider.on_changed(self.set_frame)
if self.use_blit:
self.frame_slider.drawon = False
#save background for blitting
fig = ax.figure
self.background = fig.canvas.copy_from_bbox(ax.bbox)
#enable frame scroll
fig.canvas.mpl_connect('scroll_event', self._scroll)
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#@property
#def has_coords(self):
#return self.coords is not None
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def _needs_drawing(self):
#NOTE: this method is a temp hack to return the artists that need to be
#drawn when the frame is changed (for blitting). This is in place while
#the base class is being refined.
#TODO: proper observers as modelled on draggables.machinery
needs_drawing = [self.imgplt]
if self.has_hist:
needs_drawing.extend(self.patches) #TODO: PatchCollection...
if self.autoscale:
needs_drawing.extend(self.sliders.sliders)
##[#self.imgplt.colorbar, #self.sliders.centre_knob])
return needs_drawing
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def get_data(self, i):
return self.data[i]
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def get_frame(self):
return self._frame
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#@expose.args()
def set_frame(self, i, draw=False):
'''Set frame data. draw if requested '''
i %= len(self) #wrap around! (eg. scroll past end ==> go to beginning)
i = int(round(i, 0)) #make sure we have an int
self._frame = i
data = self.get_data(i)
#ImageDisplay.draw_blit??
#set the slider axis limits
dmin, dmax = data.min(), data.max()
self.sliders.ax.set_ylim(dmin, dmax)
self.sliders.valmin, self.sliders.valmax = dmin, dmax
#needs_drawing.append()???
#set the image data
self.imgplt.set_data(data)
#needs_drawing = [self.imgplt]
if self.autoscale:
#set the slider positiions / color limits
vmin, vmax = self.get_autoscale_limits(data, autoscale=self.autoscale)
self.imgplt.set_clim(vmin, vmax)
self.sliders.set_positions((vmin, vmax))
#TODO: update hisogram values etc...
#ImageDisplay.draw_blit??
if draw:
needs_drawing = self._needs_drawing()
self.draw_blit(needs_drawing)
frame = property(get_frame, set_frame)
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def _scroll(self, event):
self.frame += [-1, +1][event.button == 'up']
self.frame_slider.set_val(self.frame)
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#@expose.args()
def draw_blit(self, artists):
#print('draw_blit')
fig = self.ax.figure
fig.canvas.restore_region(self.background)
for art in artists:
try:
self.ax.draw_artist(art)
except Exception as err:
print('drawing FAILED', art)
traceback.print_exc()
fig.canvas.blit(fig.bbox)
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
def cooDisplayFormatter(self, x, y):
s = ImageDisplay.cooDisplayFormatter(self, x,y)
return 'frame %d: %s'%(self.frame, s)
class SpecPlotter:
def __init__(self, q, freqs, x_max, hist_time, time_res, maxscale):
self.HORIZONTAL_PIXELS = len(freqs) + 1
self.q = q
self.freqs = freqs
self.x_max = x_max
self.hist_time = hist_time
self.time_res = time_res
self.maxscale = maxscale
self.hist_segments = floor(self.hist_time / self.time_res)
plt.ion()
self.figure, (self.ax, self.ax2) = plt.subplots(2)
self.figure.canvas.mpl_connect("close_event", self.exit_evt)
self.Z = np.zeros((self.hist_segments, self.HORIZONTAL_PIXELS))
self.highest_peak = 0.0 if maxscale is None else maxscale
self.scale = self.highest_peak
self.scale_slider_touched = False
plt.subplots_adjust(bottom=0.1)
self.ax_scale_slider = plt.axes([0.15, 0.03, 0.7, 0.03])
self.scale_slider = Slider(self.ax_scale_slider, "Scale", 0, self.highest_peak, self.highest_peak)
self.scale_slider.on_changed(self.scale_slider_changed)
self.X = np.linspace(self.freqs[0], self.freqs[-1], self.HORIZONTAL_PIXELS)
self.row = np.zeros(len(self.freqs))
def update(self):
while True:
while not self.q.empty():
self.row = self.q.get()
self.Z = np.vstack((self.Z, interp1d(self.freqs, self.row, kind="linear")(self.X)))
if self.Z.shape[0] > self.hist_segments:
self.Z = np.delete(self.Z, 0, axis=0)
peak_height = np.abs(self.row).max()
if self.highest_peak < peak_height:
self.highest_peak = peak_height
if self.maxscale is None:
if self.scale == self.scale_slider.valmax:
self.scale = self.highest_peak
self.ax_scale_slider.set_xlim(0, self.highest_peak)
self.scale_slider.valmax = self.highest_peak
self.scale_slider.set_val(self.scale)
self.ax.clear()
self.ax.imshow(
self.Z,
extent=(0, self.x_max, 0, self.hist_time),
cmap="Greys",
vmin=0,
vmax=self.scale,
interpolation="none",
aspect="auto",
)
self.ax.set_title("Spectrogram")
self.ax.set_xlabel("Hz")
self.ax2.clear()
self.ax2.plot(self.freqs, self.row)
self.ax2.axis([0, self.x_max, 0, self.scale])
# Draw canvas
plt.draw()
plt.pause(0.001)
def exit_evt(self, evt):
os._exit(0)
def scale_slider_changed(self, v):
self.scale = v
class viscm_editor(object):
def __init__(self, uniform_space="CAM02-UCS", min_Jp=15, max_Jp=95, xp=None, yp=None):
from .bezierbuilder import BezierModel, BezierBuilder
self._uniform_space = uniform_space
self.figure = plt.figure()
axes = _viscm_editor_axes(self.figure)
ax_btn_wireframe = plt.axes([0.7, 0.15, 0.1, 0.025])
self.btn_wireframe = Button(ax_btn_wireframe, "Show 3D gamut")
self.btn_wireframe.on_clicked(self.plot_3d_gamut)
ax_btn_wireframe = plt.axes([0.81, 0.15, 0.1, 0.025])
self.btn_save = Button(ax_btn_wireframe, "Save colormap")
self.btn_save.on_clicked(self.save_colormap)
ax_btn_props = plt.axes([0.81, 0.1, 0.1, 0.025])
self.btn_props = Button(ax_btn_props, "Properties")
self.btn_props.on_clicked(self.show_viscm)
self.prop_windows = []
axcolor = "None"
ax_jp_min = plt.axes([0.1, 0.1, 0.5, 0.03], axisbg=axcolor)
ax_jp_min.imshow(np.linspace(0, 100, 101).reshape(1, -1), cmap="gray")
ax_jp_min.set_xlim(0, 100)
ax_jp_max = plt.axes([0.1, 0.15, 0.5, 0.03], axisbg=axcolor)
ax_jp_max.imshow(np.linspace(0, 100, 101).reshape(1, -1), cmap="gray")
self.jp_min_slider = Slider(ax_jp_min, r"$J'_\mathrm{min}$", 0, 100, valinit=min_Jp)
self.jp_max_slider = Slider(ax_jp_max, r"$J'_\mathrm{max}$", 0, 100, valinit=max_Jp)
self.jp_min_slider.on_changed(self._jp_update)
self.jp_max_slider.on_changed(self._jp_update)
if xp is None:
xp = [-2.0591553836234482, 59.377014829142524, 43.552546744036135, 4.7670857511283202, -9.5059638942617539]
if yp is None:
yp = [-25.664893617021221, -21.941489361702082, 38.874113475177353, 20.567375886524871, 32.047872340425585]
self.bezier_model = BezierModel(xp, yp)
self.cmap_model = BezierCMapModel(
self.bezier_model, self.jp_min_slider.val, self.jp_max_slider.val, uniform_space
)
self.highlight_point_model = HighlightPointModel(self.cmap_model, 0.5)
self.bezier_builder = BezierBuilder(axes["bezier"], self.bezier_model)
self.bezier_gamut_viewer = GamutViewer2D(axes["bezier"], self.highlight_point_model, uniform_space)
tmp = HighlightPoint2DView(axes["bezier"], self.highlight_point_model)
self.bezier_highlight_point_view = tmp
# draw_pure_hue_angles(axes['bezier'])
axes["bezier"].set_xlim(-100, 100)
axes["bezier"].set_ylim(-100, 100)
self.cmap_view = CMapView(axes["cm"], self.cmap_model)
self.cmap_highlighter = HighlightPointBuilder(axes["cm"], self.highlight_point_model)
print("Click sliders at bottom to change min/max lightness")
print("Click on colorbar to adjust gamut view")
print("Click-drag to move control points, ")
print(" shift-click to add, control-click to delete")
def plot_3d_gamut(self, event):
fig, ax = plt.subplots(subplot_kw=dict(projection="3d"))
self.wireframe_view = WireframeView(ax, self.cmap_model, self.highlight_point_model, self._uniform_space)
plt.show()
def save_colormap(self, event):
import textwrap
template = textwrap.dedent(
"""
from matplotlib.colors import ListedColormap
from numpy import nan, inf
# Used to reconstruct the colormap in viscm
parameters = {{'xp': {xp},
'yp': {yp},
'min_Jp': {min_Jp},
'max_Jp': {max_Jp}}}
cm_data = {array_list}
test_cm = ListedColormap(cm_data, name=__file__)
if __name__ == "__main__":
import matplotlib.pyplot as plt
import numpy as np
try:
from viscm import viscm
viscm(test_cm)
except ImportError:
print("viscm not found, falling back on simple display")
plt.imshow(np.linspace(0, 100, 256)[None, :], aspect='auto',
cmap=test_cm)
plt.show()
"""
)
rgb, _ = self.cmap_model.get_sRGB(num=256)
with open("/tmp/new_cm.py", "w") as f:
array_list = np.array2string(rgb, max_line_width=78, prefix="cm_data = ", separator=",")
xp, yp = self.cmap_model.bezier_model.get_control_points()
data = dict(
array_list=array_list, xp=xp, yp=yp, min_Jp=self.cmap_model.min_Jp, max_Jp=self.cmap_model.max_Jp
)
f.write(template.format(**data))
print("*" * 50)
print("Saved colormap to /tmp/new_cm.py")
print("*" * 50)
def show_viscm(self, event):
cm = LinearSegmentedColormap.from_list("test_cm", self.cmap_model.get_sRGB(num=256)[0])
self.prop_windows.append(viscm(cm, name="test_cm"))
plt.show()
def _jp_update(self, val):
jp_min = self.jp_min_slider.val
jp_max = self.jp_max_slider.val
smallest, largest = min(jp_min, jp_max), max(jp_min, jp_max)
if (jp_min > smallest) or (jp_max < largest):
self.jp_min_slider.set_val(smallest)
self.jp_max_slider.set_val(largest)
self.cmap_model.set_Jp_minmax(smallest, largest)
class FigureContainer():
"""instantiates figure and axes, creates all other objects. Holds the
animation"""
def __init__(self, file):
#because I'm using keys that matplotlib also uses, remove some bindings
plt.rcParams['keymap.fullscreen'] = '{'
plt.rcParams['keymap.yscale'] = '}'
self.fig = plt.figure(facecolor='w')
self.fig.set_size_inches(16, 9)
g = gspec.GridSpec(5,
2,
height_ratios=[2, 6, 4, 1, 6],
width_ratios=[10, 3])
g.update(left=0.05,
right=0.95,
wspace=0.02,
hspace=.04,
bottom=.02,
top=.98)
#add the axes in the subplot areas desired
self.spatial_axis = self.fig.add_subplot(g[1, 0])
self.vid_axis = self.fig.add_subplot(g[4, 0])
self.temporal_axis = self.fig.add_subplot(g[2, 0])
#self.select_axis = self.fig.add_subplot(g[:,1])
self.sel_panel = SelectPanel(self.fig, g)
#create an axes to hold space for the buttons
#self.fig.add_subplot(g[0,0]).set_axis_off()
#add a slider
slid_ax = self.fig.add_subplot(g[3, 0])
#TODO: does this actually force it to an int
self.slide = Slider(slid_ax,
'Frame',
0,
100,
valinit=0,
valfmt='%0.0f')
self.print_manager = PrintManager(self.temporal_axis,
self.spatial_axis, self.sel_panel,
self.vid_axis, self.slide, file)
self.sel_panel.set_print_manager(self.print_manager)
self.set_slider_range()
self.prev_frame = self.slide.val
#activate pick events
self.fig.canvas.mpl_connect('pick_event', self.print_manager.on_pick)
self.fig.canvas.mpl_connect('key_press_event',
self.print_manager.on_key_press)
g2 = gspec.GridSpecFromSubplotSpec(1,
4,
subplot_spec=g[0, 0],
wspace=0.1,
hspace=0.1)
#g2.update(left=0.05, right=0.95, wspace=0.02, hspace=.04,
# bottom = .2, top = .8)
#ax=self.fig.add_subplot(g2[0,0])
#Button(ax, 'Load File').on_clicked(
# self.print_manager.initiate_split_window)
ax = self.fig.add_subplot(g2[0, 0])
self.sav_but = Button(ax, 'Save Changes')
self.sav_but.on_clicked(self.print_manager.save)
#ax=self.fig.add_subplot(g2[0,2])
#Button(ax, 'Pause Video')
#ax=self.fig.add_subplot(g2[0,3])
#Button(ax, 'View Deleted')
#and start the animation
self.anim = animation.FuncAnimation(self.fig,
self.update_func,
fargs=(),
interval=15,
repeat=True)
plt.show()
def update_func(self, j):
"""based on the value of the slider, update the video
"""
i = int(self.slide.val) + 1
if i > self.slide.valmax:
i = self.slide.valmin
self.print_manager.change_frame(self.prev_frame, i)
#self.slide.set_val(i%len(self.left_panel.frames))
self.slide.set_val(i)
self.prev_frame = i
return j
def set_slider_range(self):
"""set the slider to range between combo_prints first frame and
last frame"""
self.slide.set_val(self.print_manager.combo_prints.first_frame.min())
self.slide.valmin = self.print_manager.combo_prints.first_frame.min()
self.slide.valmax = self.print_manager.combo_prints.last_frame.max()
self.slide.ax.set_xlim(self.slide.valmin, self.slide.valmax)
class PVSlicer(object):
def __init__(self, filename, backend="Qt4Agg", clim=None):
self.filename = filename
try:
from spectral_cube import SpectralCube
cube = SpectralCube.read(filename, format='fits')
self.array = cube._data
except:
warnings.warn("spectral_cube package is not available - using astropy.io.fits directly")
from astropy.io import fits
self.array = fits.getdata(filename)
if self.array.ndim != 3:
raise ValueError("dataset does not have 3 dimensions (install the spectral_cube package to avoid this error)")
self.backend = backend
import matplotlib as mpl
mpl.use(self.backend)
import matplotlib.pyplot as plt
self.fig = plt.figure(figsize=(14, 8))
self.ax1 = self.fig.add_axes([0.1, 0.1, 0.4, 0.7])
if clim is None:
warnings.warn("clim not defined and will be determined from the data")
# To work with large arrays, sub-sample the data
# (but don't do it for small arrays)
n1 = max(self.array.shape[0] / 10, 1)
n2 = max(self.array.shape[1] / 10, 1)
n3 = max(self.array.shape[2] / 10, 1)
sub_array = self.array[::n1,::n2,::n3]
cmin = np.min(sub_array[~np.isnan(sub_array) & ~np.isinf(sub_array)])
cmax = np.max(sub_array[~np.isnan(sub_array) & ~np.isinf(sub_array)])
crange = cmax - cmin
self._clim = (cmin - crange, cmax + crange)
else:
self._clim = clim
self.slice = int(round(self.array.shape[0] / 2.))
from matplotlib.widgets import Slider
self.slice_slider_ax = self.fig.add_axes([0.1, 0.95, 0.4, 0.03])
self.slice_slider_ax.set_xticklabels("")
self.slice_slider_ax.set_yticklabels("")
self.slice_slider = Slider(self.slice_slider_ax, "3-d slice", 0, self.array.shape[0], valinit=self.slice, valfmt="%i")
self.slice_slider.on_changed(self.update_slice)
self.slice_slider.drawon = False
self.image = self.ax1.imshow(self.array[self.slice, :,:], origin='lower', interpolation='nearest', vmin=self._clim[0], vmax=self._clim[1], cmap=plt.cm.gray)
self.vmin_slider_ax = self.fig.add_axes([0.1, 0.90, 0.4, 0.03])
self.vmin_slider_ax.set_xticklabels("")
self.vmin_slider_ax.set_yticklabels("")
self.vmin_slider = Slider(self.vmin_slider_ax, "vmin", self._clim[0], self._clim[1], valinit=self._clim[0])
self.vmin_slider.on_changed(self.update_vmin)
self.vmin_slider.drawon = False
self.vmax_slider_ax = self.fig.add_axes([0.1, 0.85, 0.4, 0.03])
self.vmax_slider_ax.set_xticklabels("")
self.vmax_slider_ax.set_yticklabels("")
self.vmax_slider = Slider(self.vmax_slider_ax, "vmax", self._clim[0], self._clim[1], valinit=self._clim[1])
self.vmax_slider.on_changed(self.update_vmax)
self.vmax_slider.drawon = False
self.grid1 = None
self.grid2 = None
self.grid3 = None
self.ax2 = self.fig.add_axes([0.55, 0.1, 0.4, 0.7])
# Add slicing box
self.box = SliceCurve(colors=(0.8, 0.0, 0.0))
self.ax1.add_collection(self.box)
self.movable = MovableSliceBox(self.box, callback=self.update_pv_slice)
self.movable.connect()
# Add save button
from matplotlib.widgets import Button
self.save_button_ax = self.fig.add_axes([0.65, 0.90, 0.20, 0.05])
self.save_button = Button(self.save_button_ax, 'Save slice to FITS')
self.save_button.on_clicked(self.save_fits)
self.file_status_text = self.fig.text(0.75, 0.875, "", ha='center', va='center')
self.set_file_status(None)
self.set_file_status(None)
self.pv_slice = None
self.cidpress = self.fig.canvas.mpl_connect('button_press_event', self.click)
def set_file_status(self, status, filename=None):
if status == 'instructions':
self.file_status_text.set_text('Please enter filename in terminal')
self.file_status_text.set_color('red')
elif status == 'saved':
self.file_status_text.set_text('File successfully saved to {0}'.format(filename))
self.file_status_text.set_color('green')
else:
self.file_status_text.set_text('')
self.file_status_text.set_color('black')
self.fig.canvas.draw()
def click(self, event):
if event.inaxes != self.ax2:
return
self.slice_slider.set_val(event.ydata)
def save_fits(self, *args, **kwargs):
self.set_file_status('instructions')
print("Enter filename: ", end='')
try:
plot_name = raw_input()
except NameError:
plot_name = input()
if self.pv_slice is None:
return
try:
self.pv_slice.writeto(plot_name, overwrite=True)
except TypeError:
self.pv_slice.writeto(plot_name, clobber=True)
print("Saved file to: ", plot_name)
self.set_file_status('saved', filename=plot_name)
def update_pv_slice(self, box):
path = Path(zip(box.x, box.y))
path.width = box.width
self.pv_slice = extract_pv_slice(self.array, path)
self.ax2.cla()
self.ax2.imshow(self.pv_slice.data, origin='lower', aspect='auto', interpolation='nearest')
self.fig.canvas.draw()
def show(self, block=True):
import matplotlib.pyplot as plt
plt.show(block=block)
def update_slice(self, pos=None):
if self.array.ndim == 2:
self.image.set_array(self.array)
else:
self.slice = int(round(pos))
self.image.set_array(self.array[self.slice, :, :])
self.fig.canvas.draw()
def update_vmin(self, vmin):
if vmin > self._clim[1]:
self._clim = (self._clim[1], self._clim[1])
else:
self._clim = (vmin, self._clim[1])
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def update_vmax(self, vmax):
if vmax < self._clim[0]:
self._clim = (self._clim[0], self._clim[0])
else:
self._clim = (self._clim[0], vmax)
self.image.set_clim(*self._clim)
self.fig.canvas.draw()
def view_components(estimates, img, idx):
""" View spatial and temporal components interactively
Args:
estimates: dict
estimates dictionary contain results of VolPy
img: 2-D array
summary images for detection
idx: list
index of selected neurons
"""
n = len(idx)
fig = plt.figure(figsize=(10, 10))
axcomp = plt.axes([0.05, 0.05, 0.9, 0.03])
ax1 = plt.axes([0.05, 0.55, 0.4, 0.4])
ax3 = plt.axes([0.55, 0.55, 0.4, 0.4])
ax2 = plt.axes([0.05, 0.1, 0.9, 0.4])
s_comp = Slider(axcomp, 'Component', 0, n, valinit=0)
vmax = np.percentile(img, 98)
def arrow_key_image_control(event):
if event.key == 'left':
new_val = np.round(s_comp.val - 1)
if new_val < 0:
new_val = 0
s_comp.set_val(new_val)
elif event.key == 'right':
new_val = np.round(s_comp.val + 1)
if new_val > n:
new_val = n
s_comp.set_val(new_val)
def update(val):
i = np.int(np.round(s_comp.val))
print(f'Component:{i}')
if i < n:
ax1.cla()
imgtmp = estimates['weights'][idx][i]
ax1.imshow(imgtmp,
interpolation='None',
cmap=plt.cm.gray,
vmax=np.max(imgtmp) * 0.5,
vmin=0)
ax1.set_title(f'Spatial component {i+1}')
ax1.axis('off')
ax2.cla()
ax2.plot(estimates['t'][idx][i], alpha=0.8)
ax2.plot(estimates['t_sub'][idx][i])
ax2.plot(estimates['t_rec'][idx][i], alpha=0.4, color='red')
ax2.plot(estimates['spikes'][idx][i],
1.05 * np.max(estimates['t'][idx][i]) *
np.ones(estimates['spikes'][idx][i].shape),
color='r',
marker='.',
fillstyle='none',
linestyle='none')
ax2.set_title(f'Signal and spike times {i+1}')
ax2.legend(labels=['t', 't_sub', 't_rec', 'spikes'])
ax2.text(0.1,
0.1,
f'snr:{round(estimates["snr"][idx][i],2)}',
horizontalalignment='center',
verticalalignment='center',
transform=ax2.transAxes)
ax2.text(0.1,
0.07,
f'num_spikes: {len(estimates["spikes"][idx][i])}',
horizontalalignment='center',
verticalalignment='center',
transform=ax2.transAxes)
ax2.text(0.1,
0.04,
f'locality_test: {estimates["locality"][idx][i]}',
horizontalalignment='center',
verticalalignment='center',
transform=ax2.transAxes)
ax3.cla()
ax3.imshow(img, interpolation='None', cmap=plt.cm.gray, vmax=vmax)
imgtmp2 = imgtmp.copy()
imgtmp2[imgtmp2 == 0] = np.nan
ax3.imshow(imgtmp2,
interpolation='None',
alpha=0.5,
cmap=plt.cm.hot)
ax3.axis('off')
s_comp.on_changed(update)
s_comp.set_val(0)
fig.canvas.mpl_connect('key_release_event', arrow_key_image_control)
plt.show()
class ytViewer(object):
def __init__(self, filenames, fold=19277, nmax=100,NORM=True,dtype=int8,DEB=0,shear=0.):
self.UPDATE = False
self.color = False
self.CMAP = ['jet','seismic','Greys','plasma']
self.COLORS = ['b','g','r']
self.filenames = filenames
self.NORM = NORM
self.NMAX = nmax
self.fold = fold
self.increment = 5
self.index = 0
self.shear = shear
self.vmin = -125
self.vmax = 125
self.HORIZ_VAL = 0.05
for i in range(len(self.filenames)):
exec("self.remove_len1%d = 0" %i)
exec("self.remove_len2%d = 1" %i)
self.Y0 = 0
### To set the data arrays ###
for i in range(len(self.filenames)):
exec("self.data%d = fromfile(filenames[i],dtype=dtype)" %i)
try:
pass
except:
print '\nYou must provide existing filenames\n'
sys.exit()
self.max_index = int(len(self.data0)/self.fold)
if not(self.NMAX):
self.NMAX = self.max_index
for i in range(len(self.filenames)):
exec("self.folded_data_orig2%d = self.data%d[:self.max_index*self.fold].reshape(self.max_index,self.fold)" %(i,i))
exec("self.folded_data_orig%d = array(self.folded_data_orig2%d)" %(i,i))
exec("self.folded_data_orig3%d = array(self.folded_data_orig%d)" %(i,i))
exec("self.folded_data%d = self.folded_data_orig3%d[:self.NMAX]" %(i,i))
##################################################################
################## Start creating the figure #####################
self.fig = figure(figsize=(16,7))
if len(self.filenames)==1:
self.declare_axis_1channel()
elif len(self.filenames)==2:
self.declare_axis_2channel()
elif len(self.filenames)==3:
self.declare_axis_3channel()
for i in range(len(self.filenames)):
if not self.NORM:
exec("self.im%d = self.ax%d.imshow(self.folded_data%d, interpolation='nearest', aspect='auto',origin='lower', vmin=self.vmin, vmax=self.vmax)" %(i,i,i))
else:
exec("self.im%d = self.ax%d.imshow(self.folded_data%d, interpolation='nearest', aspect='auto',origin='lower', vmin=self.folded_data%d.min(), vmax=self.folded_data%d.max())" %(i,i,i,i,i))
for i in range(len(self.filenames)):
exec("self.cursor%d = Cursor(self.ax%d, useblit=True, color='red', linewidth=2)" %(i,i))
self.axhh = axes([0.02,0.25,0.12,0.62])
for i in range(len(self.filenames)):
exec("self.hline%d, = self.axh%d.plot(self.folded_data%d[self.Y0,:])" %(i,i,i))
exec("self.axh%d.set_xlim(0,len(self.folded_data%d[0,:]))" %(i,i))
exec("self.hhline%d, = self.axhh.plot(self.folded_data%d.mean(1),arange(self.NMAX),self.COLORS[i])" %(i,i))
self.axhh.set_ylim(0,self.NMAX-1)
if not self.NORM:
for i in range(len(self.filenames)):
exec("self.axh%d.set_ylim(self.vmin, self.vmax)" %i)
self.axhh.set_xlim(self.vmin, self.vmax)
else:
for i in range(len(self.filenames)):
exec("self.axh%d.set_ylim(self.folded_data%d.min(), self.folded_data%d.max())" %(i,i,i))
if len(self.filenames)==1:
LIM_MIN = self.folded_data0.mean(1).min()
LIM_MAX = self.folded_data0.mean(1).max()
elif len(self.filenames)==2:
LIM_MIN = min(self.folded_data0.mean(1).min(),self.folded_data1.mean(1).min())
LIM_MAX = max(self.folded_data0.mean(1).max(),self.folded_data1.mean(1).max())
elif len(self.filenames)==3:
LIM_MIN = min(self.folded_data0.mean(1).min(),self.folded_data1.mean(1).min(),self.folded_data2.mean(1).min())
LIM_MAX = max(self.folded_data0.mean(1).max(),self.folded_data1.mean(1).max(),self.folded_data2.mean(1).max())
self.axhh.set_xlim(LIM_MIN-1,LIM_MAX+1)
# create 'remove_len1' slider
for i in range(len(self.filenames)):
exec("self.remove_len1%d_slider = Slider(self.remove_len1%d_sliderax,'beg',0.,self.fold,self.remove_len1%d,'%s')" %(i,i,i,'%d'))
exec("self.remove_len1%d_slider.on_changed(self.update_tab)" %i)
# create 'remove_len2' slider
for i in range(len(self.filenames)):
exec("self.remove_len2%d_slider = Slider(self.remove_len2%d_sliderax,'end',1.,self.fold,self.remove_len2%d,'%s')" %(i,i,i,'%d'))
exec("self.remove_len2%d_slider.on_changed(self.update_tab)" %i)
# create 'index' slider
self.index_sliderax = axes([0.175,0.975,0.775,0.02])
self.index_slider = Slider(self.index_sliderax,'index',0,self.max_index-self.increment,0,'%d')
self.index_slider.on_changed(self.update_tab)
# create 'nmax' slider
self.nmax_sliderax = axes([0.175,0.955,0.775,0.02])
self.nmax_slider = Slider(self.nmax_sliderax,'nmax',0,self.max_index,self.NMAX,'%d')
self.nmax_slider.on_changed(self.update_tab)
# create 'shear' slider
self.shear_sliderax = axes([0.175,0.935,0.775,0.02])
self.shear_slider = Slider(self.shear_sliderax,'Shear',-0.5,0.5,self.shear,'%1.2f')
self.shear_slider.on_changed(self.update_shear)
cid = self.fig.canvas.mpl_connect('motion_notify_event', self.mousemove)
cid2 = self.fig.canvas.mpl_connect('key_press_event', self.keypress)
VERT_VAL = -4
font0 = FontProperties()
font1 = font0.copy()
font1.set_weight('bold')
mpl.pyplot.text(-0.72,-32+VERT_VAL,'Useful keys:',fontsize=18,fontproperties=font1)
mpl.pyplot.text(-0.72,-41+VERT_VAL,'"c" to change colormap\n "v" to change vertical\n /colorscale\n " " to pause\n "w"/"x" set Ch1 REMOVE\n sliders values to Ch2/3\n "t" Retrigger mode\n (NOT TOO MUCH POINTS) \n "q" to exit',fontsize=18)
self.axe_toggledisplay = self.fig.add_axes([0.,0.,1.0,0.02])
if self.UPDATE:
self.plot_circle(0,0,2,fc='#00FF7F')
else:
self.plot_circle(0,0,2,fc='#FF4500')
mpl.pyplot.axis('off')
gobject.idle_add(self.update_plot)
show()
### BEGIN main loop ###
def update_plot(self):
while self.UPDATE:
### Compute the array to plot ###
for i in range(len(self.filenames)):
exec("self.folded_data%d = self.folded_data_orig3%d[self.index:(self.index+self.NMAX)]" %(i,i))
### Update picture ###
for i in range(len(self.filenames)):
exec("self.im%d.set_data(self.folded_data%d)" %(i,i))
exec("self.hline%d.set_ydata(self.folded_data%d[self.Y0,:])" %(i,i))
exec("self.hhline%d.set_xdata(self.folded_data%d.mean(1))" %(i,i))
self.index = self.index + self.increment
self.index_slider.set_val(self.index)
self.fig.canvas.draw()
return True
return False
### END main loop ###
def update_tab(self,val):
for i in range(len(self.filenames)):
exec("self.remove_len1%d = int(self.remove_len1%d_slider.val)" %(i,i))
exec("self.remove_len2%d = int(self.remove_len2%d_slider.val)" %(i,i))
self.index = int(round(self.index_slider.val,0))
self.NMAX = int(round(self.nmax_slider.val,0))
self.Y0 = 0
self.update_tabs()
self.norm_fig()
self.fig.canvas.draw()
def update_tabs(self):
for i in range(len(self.filenames)):
exec("self.folded_data_orig3%d = array(self.folded_data_orig%d[:,self.remove_len1%d:-self.remove_len2%d])" %(i,i,i,i))
exec("self.process_data(self.shear,%d)" %i)
exec("self.folded_data%d = self.folded_data_orig3%d[self.index:(self.index+self.NMAX)]" %(i,i))
def process_data(self,val,i):
""" Redress data in the space/time diagram """
exec("dd = self.folded_data_orig2%d.copy()" %i)
for k in range(0,dd.shape[0]):
exec("dd[k,:] = roll(self.folded_data_orig2%d[k,:], int(k*val))" %i)
exec("self.folded_data_orig%d = dd" %i)
def norm_fig(self):
if len(self.filenames)==1:
LIM_MIN = self.folded_data0.mean(1).min()
LIM_MAX = self.folded_data0.mean(1).max()
elif len(self.filenames)==2:
LIM_MIN = min(self.folded_data0.mean(1).min(),self.folded_data1.mean(1).min())
LIM_MAX = max(self.folded_data0.mean(1).max(),self.folded_data1.mean(1).max())
elif len(self.filenames)==3:
LIM_MIN = min(self.folded_data0.mean(1).min(),self.folded_data1.mean(1).min(),self.folded_data2.mean(1).min())
LIM_MAX = max(self.folded_data0.mean(1).max(),self.folded_data1.mean(1).max(),self.folded_data2.mean(1).max())
self.axhh.clear()
if not self.NORM:
for i in range(len(self.filenames)):
exec("self.ax%d.clear()" %i)
exec("self.im%d = self.ax%d.imshow(self.folded_data%d,cmap=self.CMAP[0], interpolation='nearest', aspect='auto',origin='lower', vmin=self.vmin, vmax=self.vmax)" %(i,i,i))
exec("self.axh%d.clear()" %i)
exec("self.hline%d, = self.axh%d.plot(self.folded_data%d[self.Y0,:])" %(i,i,i))
exec("self.axh%d.set_ylim(self.vmin, self.vmax)" %i)
exec("self.axh%d.set_xlim(0, len(self.folded_data%d[0]))" %(i,i))
exec("self.hhline%d, = self.axhh.plot(self.folded_data%d.mean(1),arange(len(self.folded_data%d.mean(1))),self.COLORS[i])" %(i,i,i))
self.axhh.set_ylim(0,self.max_index-self.index-1)
self.axhh.set_xlim(self.vmin, self.vmax)
else:
for i in range(len(self.filenames)):
exec("self.ax%d.clear()" %i)
exec("self.im%d = self.ax%d.imshow(self.folded_data%d,cmap=self.CMAP[0], interpolation='nearest', aspect='auto',origin='lower', vmin=self.folded_data%d.min(), vmax=self.folded_data%d.max())" %(i,i,i,i,i))
exec("self.axh%d.clear()" %i)
exec("self.hline%d, = self.axh%d.plot(self.folded_data%d[self.Y0,:])" %(i,i,i))
exec("self.axh%d.set_ylim(self.folded_data%d.min(), self.folded_data%d.max())" %(i,i,i))
exec("self.axh%d.set_xlim(0, len(self.folded_data%d[0]))" %(i,i))
exec("self.hhline%d, = self.axhh.plot(self.folded_data%d.mean(1),arange(len(self.folded_data%d.mean(1))),self.COLORS[i])" %(i,i,i))
self.axhh.set_ylim(0,len(self.folded_data0.mean(1)))
self.axhh.set_xlim(LIM_MIN-1,LIM_MAX+1)
self.fig.canvas.draw()
### BEGIN Slider actions ###
def update_shear(self,val):
self.shear = round(self.shear_slider.val,2)
self.update_tabs()
self.update_tab(0)
self.norm_fig()
self.fig.canvas.draw()
def update_cut(self):
for i in range(len(self.filenames)):
exec("self.hline%d.set_ydata(self.folded_data%d[self.Y0,:])" %(i,i))
self.fig.canvas.draw()
### END Slider actions ###
### BEGIN actions to the window ###
def toggle_update(self):
self.UPDATE = not(self.UPDATE)
if self.UPDATE:
gobject.idle_add(self.update_plot)
self.color = not(self.color)
if not(self.color):
self.patch.remove()
self.axe_toggledisplay = self.fig.add_axes([0.,0.,1.0,0.02])
self.axe_toggledisplay.clear()
self.plot_circle(0,0,2,fc='#FF4500')
mpl.pyplot.axis('off')
self.fig.canvas.draw()
else:
self.patch.remove()
self.axe_toggledisplay = self.fig.add_axes([0.,0.,1.0,0.02])
self.axe_toggledisplay.clear()
self.plot_circle(0,0,2,fc='#00FF7F')
mpl.pyplot.axis('off')
self.fig.canvas.draw()
def keypress(self, event):
if event.key == 'q': # eXit
del event
sys.exit()
elif event.key=='c':
del event
self.CMAP = roll(self.CMAP,-1)
self.norm_fig()
self.fig.canvas.draw()
elif event.key=='v':
del event
self.NORM = not(self.NORM)
self.norm_fig()
elif event.key == ' ': # play/pause
self.toggle_update()
elif event.key == 'w':
if len(self.filenames)>=2:
print 'Set REMOVE values of channel 1 to channel 2'
self.remove_len11_slider.set_val(self.remove_len10)
self.remove_len21_slider.set_val(self.remove_len20)
elif event.key == 'x':
if len(self.filenames)>=3:
print 'Set REMOVE values of channel 1 to channel 3'
self.remove_len12_slider.set_val(self.remove_len10)
self.remove_len22_slider.set_val(self.remove_len20)
elif event.key == 't':
print 'Trying to smooth from index',self.index
self.smooth_array()
print 'Done MF'
else:
print 'Key '+str(event.key)+' not known'
def mousemove(self, event):
# called on each mouse motion to get mouse position
if len(self.filenames)==1:
if event.inaxes!=self.ax0: return
elif len(self.filenames)==2:
if event.inaxes!=self.ax0 and event.inaxes!=self.ax1: return
elif len(self.filenames)==3:
if event.inaxes!=self.ax0 and event.inaxes!=self.ax1 and event.inaxes!=self.ax2: return
self.X0 = int(round(event.xdata,0))
self.Y0 = int(round(event.ydata,0))
self.update_cut()
### END actions to the window ###
### Divers useful functions ###
def plot_circle(self,x,y,r,fc='r'):
"""Plot a circle of radius r at position x,y"""
cir = mpl.patches.Circle((x,y), radius=r, fc=fc)
self.patch = mpl.pyplot.gca().add_patch(cir)
def smooth_array(self):
if len(self.filenames)>=2 and self.UPDATE==False:
self.fig2 = figure(5,figsize=(16,7))
clf()
for i in range(len(self.filenames)):
exec("self.folded_data%d = self.folded_data_orig3%d[self.index:(self.index+self.NMAX)]" %(i,i))
if len(self.filenames)==2:
self.folded_data_retriggered1,self.folded_data_retriggered0 = self.trig_by_interpolation_pola(self.folded_data1,self.folded_data0)
self.fig2ax0 = self.fig2.add_subplot(111)
self.fig2ax0.clear()
self.fig2ax0.imshow(self.folded_data_retriggered0,cmap=self.CMAP[0], interpolation='nearest', aspect='auto',origin='lower', vmin=self.folded_data_retriggered0.min(), vmax=self.folded_data_retriggered0.max())
elif len(self.filenames)==3:
self.folded_data_retriggered2,self.folded_data_retriggered0,self.folded_data_retriggered1 = self.trig_by_interpolation_pola(self.folded_data2,self.folded_data0,self.folded_data1)
self.fig2ax1 = self.fig2.add_subplot(121)
self.fig2ax1.clear()
self.fig2ax1.imshow(self.folded_data_retriggered0,cmap=self.CMAP[0], interpolation='nearest', aspect='auto',origin='lower', vmin=self.folded_data_retriggered0.min(), vmax=self.folded_data_retriggered0.max())
self.fig2ax2 = self.fig2.add_subplot(122)
self.fig2ax2.clear()
self.fig2ax2.imshow(self.folded_data_retriggered1,cmap=self.CMAP[0], interpolation='nearest', aspect='auto',origin='lower', vmin=self.folded_data_retriggered1.min(), vmax=self.folded_data_retriggered1.max())
show(False)
self.fig2.canvas.draw()
else:
print 'This function REQUIRES a trigger'
def trig_by_interpolation_pola(self,data,pola1,pola2=None,thr=30,FACT=25,num_trig=0,DOWN=False):
if DOWN:
pos = self.find_down(data,thr)
else:
pos = self.find_up(data,thr)
l = data # for a first trigg -> #array([data[pos[i]-100:pos[i]+100] for i in xrange(len(pos)-1)])
lp1 = pola1
if pola2 is not None: lp2 = pola2
ll = []
trace = []
trace_p1 = []
if pola2 is not None: trace_p2 = []
for i in range(len(l)):
b = interpolate.interp1d(linspace(0,len(l[i]),len(l[i])),l[i])
bp1 = interpolate.interp1d(linspace(0,len(lp1[i]),len(lp1[i])),lp1[i])
if pola2 is not None: bp2 = interpolate.interp1d(linspace(0,len(lp2[i]),len(lp2[i])),lp2[i])
xnew2 = linspace(0, len(l[i]),FACT*len(l[i]))
xnew = linspace(0, len(lp1[i]),FACT*len(lp1[i]))
xnew3 = linspace(0, len(lp2[i]),FACT*len(lp2[i]))
try:
temp2 = b(xnew2)
temp2_p1 = bp1(xnew)
if pola2 is not None: temp2_p2 = bp2(xnew3)
if DOWN:
temp = self.find_down(temp2,thr)
else:
temp = self.find_up(temp2,thr)
ll.append(temp[num_trig]) # If several downward event found for the trigger
trace.append(temp2)
trace_p1.append(temp2_p1)
if pola2 is not None: trace_p2.append(temp2_p2)
except:
print '%d WARNING: Error repering => skiping a line'%i
lll = []
lllp1 = []
if pola2 is not None: lllp2 = []
ll = array(ll)-array(ll).min()
for i in range(len(ll)):
lll.append(roll(trace[i],-ll[i]))
lllp1.append(roll(trace_p1[i],-ll[i]))
if pola2 is not None: lllp2.append(roll(trace_p2[i],-ll[i]))
return (array(lll),array(lllp1),array(lllp2)) if pola2 is not None else (array(lll),array(lllp1))
def find_down(self,d, threshold):
digitized = zeros(shape=d.shape, dtype=uint8)
digitized[where(d < threshold)] = 255
derivative = digitized[1:]-digitized[0:-1]
indices = where(derivative == 255)[0]
return indices
def find_up(self,d, threshold):
digitized = zeros(shape=d.shape, dtype=uint8)
digitized[where(d > threshold)] = 255
derivative = digitized[1:]-digitized[0:-1]
indices = where(derivative == 255)[0]
return indices
def declare_axis_1channel(self):
self.ax0 = axes([0.125+self.HORIZ_VAL,0.25,0.81,0.62])
self.axh0 = axes([0.125+self.HORIZ_VAL,0.05,0.81,0.15])
self.remove_len10_sliderax = axes([0.125+self.HORIZ_VAL,0.91,0.78,0.02])
self.remove_len20_sliderax = axes([0.125+self.HORIZ_VAL,0.88,0.78,0.02])
def declare_axis_2channel(self):
self.ax0 = axes([0.125+self.HORIZ_VAL,0.25,0.395,0.62])
self.ax1 = axes([0.54+self.HORIZ_VAL,0.25,0.395,0.62])
self.axh0 = axes([0.125+self.HORIZ_VAL,0.05,0.395,0.15])
self.axh1 = axes([0.54+self.HORIZ_VAL,0.05,0.395,0.15])
self.remove_len10_sliderax = axes([0.125+self.HORIZ_VAL,0.91,0.37,0.02])
self.remove_len11_sliderax = axes([0.54+self.HORIZ_VAL,0.91,0.37,0.02])
self.remove_len20_sliderax = axes([0.125+self.HORIZ_VAL,0.88,0.37,0.02])
self.remove_len21_sliderax = axes([0.54+self.HORIZ_VAL,0.88,0.37,0.02])
def declare_axis_3channel(self):
self.ax0 = axes([0.125+self.HORIZ_VAL,0.25,0.25,0.62])
self.ax1 = axes([0.405+self.HORIZ_VAL,0.25,0.25,0.62])
self.ax2 = axes([0.685+self.HORIZ_VAL,0.25,0.25,0.62])
self.axh0 = axes([0.125+self.HORIZ_VAL,0.05,0.25,0.15])
self.axh1 = axes([0.405+self.HORIZ_VAL,0.05,0.25,0.15])
self.axh2 = axes([0.685+self.HORIZ_VAL,0.05,0.25,0.15])
self.remove_len10_sliderax = axes([0.125+self.HORIZ_VAL,0.91,0.25,0.02])
self.remove_len11_sliderax = axes([0.405+self.HORIZ_VAL,0.91,0.25,0.02])
self.remove_len12_sliderax = axes([0.685+self.HORIZ_VAL,0.91,0.25,0.02])
self.remove_len20_sliderax = axes([0.125+self.HORIZ_VAL,0.88,0.25,0.02])
self.remove_len21_sliderax = axes([0.405+self.HORIZ_VAL,0.88,0.25,0.02])
self.remove_len22_sliderax = axes([0.685+self.HORIZ_VAL,0.88,0.25,0.02])
class timeseriesViewer():
"""Class for tsview.py
Example:
cmd = 'tsview.py timeseries_ERA5_ramp_demErr.h5'
obj = timeseriesViewer(cmd)
obj.configure()
obj.plot()
"""
def __init__(self, cmd=None, iargs=None):
if cmd:
iargs = cmd.split()[1:]
self.cmd = cmd
self.iargs = iargs
# print command line
cmd = '{} '.format(os.path.basename(__file__))
cmd += ' '.join(iargs)
print(cmd)
# figure variables
self.figname_img = 'Cumulative Displacement Map'
self.figsize_img = None
self.fig_img = None
self.ax_img = None
self.cbar_img = None
self.img = None
self.ax_tslider = None
self.tslider = None
self.figname_pts = 'Point Displacement Time-series'
self.figsize_pts = None
self.fig_pts = None
self.ax_pts = None
return
def configure(self):
inps = cmd_line_parse(self.iargs)
inps, self.atr = read_init_info(inps)
# copy inps to self object
for key, value in inps.__dict__.items():
setattr(self, key, value)
# input figsize for the point time-series plot
self.figsize_pts = self.fig_size
self.pts_marker = 'r^'
self.pts_marker_size = 6.
return
def plot(self):
# read 3D time-series
self.ts_data, self.mask = read_timeseries_data(self)[0:2]
# Figure 1 - Cumulative Displacement Map
self.fig_img = plt.figure(self.figname_img, figsize=self.figsize_img)
# Figure 1 - Axes 1 - Displacement Map
self.ax_img = self.fig_img.add_axes([0.125, 0.25, 0.75, 0.65])
img_data = np.array(
self.ts_data[0][self.idx, :, :]) ####################
img_data[self.mask == 0] = np.nan
self.plot_init_image(img_data)
# Figure 1 - Axes 2 - Time Slider
self.ax_tslider = self.fig_img.add_axes([0.2, 0.1, 0.6, 0.07])
self.plot_init_time_slider(init_idx=self.idx, ref_idx=self.ref_idx)
self.tslider.on_changed(self.update_time_slider)
# Figure 2 - Time Series Displacement - Point
self.fig_pts, self.ax_pts = plt.subplots(num=self.figname_pts,
figsize=self.figsize_pts)
if self.yx:
d_ts = self.plot_point_timeseries(self.yx)
# Output
if self.save_fig:
save_ts_plot(self.yx, self.fig_img, self.fig_pts, d_ts, self)
# Final linking of the canvas to the plots.
self.fig_img.canvas.mpl_connect('button_press_event',
self.update_plot_timeseries)
self.fig_img.canvas.mpl_connect('key_press_event', self.on_key_event)
if self.disp_fig:
vprint('showing ...')
msg = '\n------------------------------------------------------------------------'
msg += '\nTo scroll through the image sequence:'
msg += '\n1) Move the slider, OR'
msg += '\n2) Press left or right arrow key (if not responding, click the image and try again).'
msg += '\n------------------------------------------------------------------------'
vprint(msg)
plt.show()
return
def plot_init_image(self, img_data):
# prepare data
if self.wrap:
if self.disp_unit_img == 'radian':
img_data *= self.range2phase
img_data = ut.wrap(img_data, wrap_range=self.wrap_range)
# Title and Axis Label
disp_date = self.dates[self.idx].strftime('%Y-%m-%d')
self.fig_title = 'N = {}, Time = {}'.format(self.idx, disp_date)
# Initial Pixel of interest
self.pts_yx = None
self.pts_lalo = None
if self.yx and self.yx != self.ref_yx:
self.pts_yx = np.array(self.yx).reshape(-1, 2)
if self.lalo:
self.pts_lalo = np.array(self.lalo).reshape(-1, 2)
# call view.py to plot
self.img, self.cbar_img = view.plot_slice(self.ax_img, img_data,
self.atr, self)[2:4]
return self.img, self.cbar_img
def plot_init_time_slider(self, init_idx=-1, ref_idx=0):
val_step = np.min(np.diff(self.yearList))
val_min = self.yearList[0]
val_max = self.yearList[-1]
self.tslider = Slider(self.ax_tslider,
label='Years',
valinit=self.yearList[init_idx],
valmin=val_min,
valmax=val_max,
valstep=val_step)
bar_width = val_step / 4.
datex = np.array(self.yearList) - bar_width / 2.
self.tslider.ax.bar(datex,
np.ones(len(datex)),
bar_width,
facecolor='black',
ecolor=None)
self.tslider.ax.bar(datex[ref_idx],
1.,
bar_width * 3,
facecolor='crimson',
ecolor=None)
# xaxis tick format
if np.floor(val_max) == np.floor(val_min):
digit = 10.
else:
digit = 1.
self.tslider.ax.set_xticks(
np.round(np.linspace(val_min, val_max, num=5) * digit) / digit)
self.tslider.ax.xaxis.set_minor_locator(MultipleLocator(1. / 12.))
self.tslider.ax.set_xlim([val_min, val_max])
self.tslider.ax.set_yticks([])
return self.tslider
def update_time_slider(self, val):
"""Update Displacement Map using Slider"""
idx = np.argmin(np.abs(np.array(self.yearList) - self.tslider.val))
# update title
disp_date = self.dates[idx].strftime('%Y-%m-%d')
self.ax_img.set_title('N = {n}, Time = {t}'.format(n=idx, t=disp_date),
fontsize=self.font_size)
# read data
data_img = np.array(self.ts_data[0][idx, :, :])
data_img[self.mask == 0] = np.nan
if self.wrap:
if self.disp_unit_img == 'radian':
data_img *= self.range2phase
data_img = ut.wrap(data_img, wrap_range=self.wrap_range)
# update data
self.img.set_data(data_img)
self.idx = idx
self.fig_img.canvas.draw()
return
def plot_point_timeseries(self, yx):
"""Plot point displacement time-series at pixel [y, x]
Parameters: yx : list of 2 int
Returns: d_ts : 2D np.array in size of (num_date, num_file)
"""
self.ax_pts.cla()
# plot scatter in different size for different files
num_file = len(self.ts_data)
if num_file <= 2: ms_step = 4
elif num_file == 3: ms_step = 3
elif num_file == 4: ms_step = 2
elif num_file >= 5: ms_step = 1
d_ts = []
y = yx[0] - self.pix_box[1]
x = yx[1] - self.pix_box[0]
for i in range(num_file - 1, -1, -1):
# get displacement data
d_tsi = self.ts_data[i][:, y, x]
if self.zero_first:
d_tsi -= d_tsi[self.zero_idx]
d_ts.append(d_tsi)
# get plot parameter - namespace ppar
ppar = argparse.Namespace()
ppar.label = self.file_label[i]
ppar.ms = self.marker_size - ms_step * (num_file - 1 - i)
ppar.mfc = pp.mplColors[num_file - 1 - i]
if self.mask[y, x] == 0:
ppar.mfc = 'gray'
if self.offset:
d_tsi += self.offset * (num_file - 1 - i)
# plot
if not np.all(np.isnan(d_tsi)):
self.ax_pts = self.ts_plot_func(self.ax_pts, d_tsi, self, ppar)
# axis format
self.ax_pts = _adjust_ts_axis(self.ax_pts, self)
title_ts = _get_ts_title(yx[0], yx[1], self.coord)
if self.mask[y, x] == 0:
title_ts += ' (masked out)'
if self.disp_title:
self.ax_pts.set_title(title_ts, fontsize=self.font_size)
if self.tick_right:
self.ax_pts.yaxis.tick_right()
self.ax_pts.yaxis.set_label_position("right")
# legend
if len(self.ts_data) > 1:
self.ax_pts.legend()
# Print to terminal
vprint('\n---------------------------------------')
vprint(title_ts)
float_formatter = lambda x: [float('{:.2f}'.format(i)) for i in x]
vprint(float_formatter(d_ts[0]))
if not np.all(np.isnan(d_ts[0])):
# stat info
vprint('displacement range: [{:.2f}, {:.2f}] {}'.format(
np.nanmin(d_ts[0]), np.nanmax(d_ts[0]), self.disp_unit))
# estimate (print) slope
estimate_slope(d_ts[0],
self.yearList,
ex_flag=self.ex_flag,
disp_unit=self.disp_unit)
# update figure
self.fig_pts.canvas.draw()
return d_ts
def update_plot_timeseries(self, event):
"""Event function to get y/x from button press"""
if event.inaxes == self.ax_img:
# get row/col number
if self.fig_coord == 'geo':
y, x = self.coord.geo2radar(event.ydata,
event.xdata,
print_msg=False)[0:2]
else:
y, x = int(event.ydata + 0.5), int(event.xdata + 0.5)
# plot time-series displacement
self.plot_point_timeseries((y, x))
return
def on_key_event(self, event):
"""Slide images with left/right key on keyboard"""
if event.inaxes and event.inaxes.figure == self.fig_img:
idx = None
if event.key == 'left':
idx = max(self.idx - 1, 0)
elif event.key == 'right':
idx = min(self.idx + 1, self.num_date - 1)
if idx is not None and idx != self.idx:
# update title
disp_date = self.dates[idx].strftime('%Y-%m-%d')
self.ax_img.set_title('N = {n}, Time = {t}'.format(
n=idx, t=disp_date),
fontsize=self.font_size)
# read data
data_img = np.array(self.ts_data[0][idx, :, :])
data_img[self.mask == 0] = np.nan
if self.wrap:
if self.disp_unit_img == 'radian':
data_img *= self.range2phase
data_img = ut.wrap(data_img, wrap_range=self.wrap_range)
# update
self.img.set_data(data_img) # update image
self.tslider.set_val(self.yearList[idx]) # update slider
self.idx = idx
self.fig_img.canvas.draw()
return
def profileview(model):
plt.rc('font', size=8)
fig, axs = plt.subplots(3, sharex=True)
axs[0].set_title('protonic potential (V)')
axs[1].set_title('oxygen molar fraction')
axs[2].set_title('current density (A/m2)')
axs[2].set_xlabel('distance from membrane (um)')
fig.subplots_adjust(right=0.45)
# the mask is assumed to account for boundary values
mask = ~(model.gdl | model.membrane)
# we denote the distance from the membrane wall x
x = model.distance_from_membrane[mask] * 1E6
# the polarization curve is permanent and unique
pcax = fig.add_axes([.5, .2, .4, .7])
pcax.set_title('polarization curve')
pcax.yaxis.tick_right()
pcax.yaxis.set_label_position("right")
pcax.set_xlabel('geometric current density (A/cm**2)')
pcax.set_ylabel('voltage at gdl (V)')
polcurve, = pcax.plot([], [], 'ko-')
def update(V):
# try networks generate l==1
model.resolve(V, 263, flood=False)
i = model.current_history[-1][mask]
I = i.sum() / model.face_area / 1000
reading = (I, V)
insert_point(reading, polcurve)
pcax.relim()
pcax.autoscale_view()
for ax, yh in zip(axs,
[ model.proton_history,
model.oxygen_history,
model.current_history,
]):
y = yh[-1]
update_ax(ax, x, y)
fig.canvas.draw()
slider = Slider(label='V',
ax=fig.add_axes([.5, .05, .4, .03]),
valmin=0,
valmax=1.2,
)
slider.on_changed(update)
# generate an undersampled polcurve
for V in np.linspace(0.05, 1.0, 10):
slider.set_val(V)
slider.set_val(0.65)
# some extra interactivity
fig.canvas.mpl_connect('button_press_event',
lambda event: slider.set_val(event.ydata)
if pcax is event.inaxes else None)
plt.show()
def view_patches_bar(Yr, A, C, b, f, d1, d2, YrA=None, img=None):
"""view spatial and temporal components interactively
Parameters:
-----------
Yr: np.ndarray
movie in format pixels (d) x frames (T)
A: sparse matrix
matrix of spatial components (d x K)
C: np.ndarray
matrix of temporal components (K x T)
b: np.ndarray
spatial background (vector of length d)
f: np.ndarray
temporal background (vector of length T)
d1,d2: np.ndarray
frame dimensions
YrA: np.ndarray
ROI filtered residual as it is given from update_temporal_components
If not given, then it is computed (K x T)
img: np.ndarray
background image for contour plotting. Default is the image of all spatial components (d1 x d2)
"""
pl.ion()
if 'csc_matrix' not in str(type(A)):
A = csc_matrix(A)
if 'array' not in str(type(b)):
b = b.toarray()
nr, T = C.shape
nb = f.shape[0]
nA2 = np.sqrt(np.array(A.power(2).sum(axis=0))).squeeze()
if YrA is None:
Y_r = spdiags(old_div(1, nA2), 0, nr, nr) * (A.T.dot(Yr) -
(A.T.dot(b)).dot(f) - (A.T.dot(A)).dot(C)) + C
else:
Y_r = YrA + C
if img is None:
img = np.reshape(np.array(A.mean(axis=1)), (d1, d2), order='F')
fig = pl.figure(figsize=(10, 10))
axcomp = pl.axes([0.05, 0.05, 0.9, 0.03])
ax1 = pl.axes([0.05, 0.55, 0.4, 0.4])
ax3 = pl.axes([0.55, 0.55, 0.4, 0.4])
ax2 = pl.axes([0.05, 0.1, 0.9, 0.4])
s_comp = Slider(axcomp, 'Component', 0, nr + nb - 1, valinit=0)
vmax = np.percentile(img, 95)
def update(val):
i = np.int(np.round(s_comp.val))
print(('Component:' + str(i)))
if i < nr:
ax1.cla()
imgtmp = np.reshape(A[:, i].toarray(), (d1, d2), order='F')
ax1.imshow(imgtmp, interpolation='None', cmap=pl.cm.gray, vmax=np.max(imgtmp)*0.5)
ax1.set_title('Spatial component ' + str(i + 1))
ax1.axis('off')
ax2.cla()
ax2.plot(np.arange(T), Y_r[i], 'c', linewidth=3)
ax2.plot(np.arange(T), C[i], 'r', linewidth=2)
ax2.set_title('Temporal component ' + str(i + 1))
ax2.legend(labels=['Filtered raw data', 'Inferred trace'])
ax3.cla()
ax3.imshow(img, interpolation='None', cmap=pl.cm.gray, vmax=vmax)
imgtmp2 = imgtmp.copy()
imgtmp2[imgtmp2 == 0] = np.nan
ax3.imshow(imgtmp2, interpolation='None',
alpha=0.5, cmap=pl.cm.hot)
ax3.axis('off')
else:
ax1.cla()
bkgrnd = np.reshape(b[:, i - nr], (d1, d2), order='F')
ax1.imshow(bkgrnd, interpolation='None')
ax1.set_title('Spatial background ' + str(i + 1 - nr))
ax1.axis('off')
ax2.cla()
ax2.plot(np.arange(T), np.squeeze(np.array(f[i - nr, :])))
ax2.set_title('Temporal background ' + str(i + 1 - nr))
def arrow_key_image_control(event):
if event.key == 'left':
new_val = np.round(s_comp.val - 1)
if new_val < 0:
new_val = 0
s_comp.set_val(new_val)
elif event.key == 'right':
new_val = np.round(s_comp.val + 1)
if new_val > nr + nb:
new_val = nr + nb
s_comp.set_val(new_val)
else:
pass
s_comp.on_changed(update)
s_comp.set_val(0)
fig.canvas.mpl_connect('key_release_event', arrow_key_image_control)
pl.show()
class Hist4D(object):
def __init__(self, save_only=False):
self.fig=None
self.cubes_info=None
self.slow=None
self.shigh=None
self.colormap=None
self.save_only = save_only
def draw_cubes(self,_axes, vals, edges):
'''
ax=Axes3D handle
edges=matrix L+1xM+1xN+1 result of histogramdd
vals=matrix LxMxN result of histogramdd
colormap=color map to be matched with nonzero vals
'''
edx, edy, edz = np.meshgrid(edges[0], edges[1], edges[2])
edx_rolled = np.roll(edx, -1, axis=1)
edy_rolled = np.roll(edy, -1, axis=0)
edz_rolled = np.roll(edz, -1, axis=2)
edx_rolled = edx_rolled[:-1, :-1, :-1].ravel()
edy_rolled = edy_rolled[:-1, :-1, :-1].ravel()
edz_rolled = edz_rolled[:-1, :-1, :-1].ravel()
edx = edx[:-1, :-1, :-1].ravel()
edy = edy[:-1, :-1, :-1].ravel()
edz = edz[:-1, :-1, :-1].ravel()
vals = vals.ravel()
vdraw_cube = np.vectorize(self.draw_cube, excluded='_axes')
cubes_handles = vdraw_cube(_axes, edx[vals>0],
edx_rolled[vals>0],
edy[vals>0],
edy_rolled[vals>0],
edz[vals > 0],
edz_rolled[vals > 0],
vals[vals>0]/float(np.max(vals)))
cubes_data = [a for a in zip(vals[vals>0],cubes_handles)]
self.cubes_info=dict()
for k, v in cubes_data:
self.cubes_info[k] = self.cubes_info.get(k, ()) + tuple(v) #+(v,)
def set_sliders(self,splot1,splot2):
maxlim=max(self.cubes_info.keys())
axcolor = 'lightgoldenrodyellow'
#low_vis = self.fig.add_axes([0.25, 0.1, 0.65, 0.03], axisbg=axcolor)
#high_vis = self.fig.add_axes([0.25, 0.15, 0.65, 0.03], axisbg=axcolor)
self.slow = Slider(splot1,'low', 0.0, maxlim, valfmt='%0.0f')
self.shigh = Slider(splot2, 'high', 0.0 , maxlim, valfmt='%0.0f')
self.slow.on_changed(self.update)
self.shigh.on_changed(self.update)
self.slow.set_val(0)
self.shigh.set_val(maxlim)
def update(self,val):
visible = [(k,v) for k, v in self.cubes_info.items() if k >
self.slow.val and k<=
self.shigh.val]
invisible = [v for k, v in self.cubes_info.items() if k <=
self.slow.val or k>
self.shigh.val]
for (k,sublist) in visible:
for item in sublist:
print item.set_alpha
item.set_alpha(k)
for item in [item for sublist in invisible for item in sublist]:
item.set_alpha(0)
total=[v for k,v in self.cubes_info.items()]
self.fig.canvas.draw_idle()
def draw_cube(self,_axes, x1_coord, x2_coord,
y1_coord, y2_coord,
z1_coord, z2_coord,
color_ind):
'''
draw a cube given cube limits and color
'''
_x_coord, _y_coord, _z_coord = np.meshgrid([x1_coord, x2_coord],
[y1_coord, y2_coord],
[z1_coord, z2_coord])
tmp1 = np.concatenate((_x_coord.ravel()[None, :], _y_coord.ravel()[
None, :], _z_coord.ravel()[None, :]), axis=0)
tmp2 = tmp1.copy()
tmp2[:, [0, 1]], tmp2[:, [6, 7]] = tmp2[
:, [6, 7]].copy(), tmp2[:, [0, 1]].copy()
tmp3 = tmp2.copy()
tmp3[:, [0, 2]], tmp3[:, [5, 7]] = tmp3[
:, [5, 7]].copy(), tmp3[:, [0, 2]].copy()
points = np.concatenate((tmp1, tmp2, tmp3), axis=1)
points = points.T.reshape(6, 4, 3)
'''
collection = Poly3DCollection(points,
facecolors=self.colormap(float(color_ind)),
linewidths=0
)
_axes.add_collection3d(collection)
return collection
'''
surf = []
for count in range(6):
surf.append(_axes.plot_surface(points[count, :, 0].reshape(2, 2),
points[count, :, 1].reshape(2, 2),
points[count, :, 2].reshape(2, 2),
color=self.colormap(float(color_ind)),
linewidth=0,
antialiased=True,
shade=False))
return surf
def array2cmap(self,X):
N = X.shape[0]
r = np.linspace(0., 1., N+1)
r = np.sort(np.concatenate((r, r)))[1:-1]
rd = np.concatenate([[X[i, 0], X[i, 0]] for i in xrange(N)])
gr = np.concatenate([[X[i, 1], X[i, 1]] for i in xrange(N)])
bl = np.concatenate([[X[i, 2], X[i, 2]] for i in xrange(N)])
al = np.concatenate([[X[i, 3], X[i, 3]] for i in xrange(N)])
rd = tuple([(r[i], rd[i], rd[i]) for i in xrange(2 * N)])
gr = tuple([(r[i], gr[i], gr[i]) for i in xrange(2 * N)])
bl = tuple([(r[i], bl[i], bl[i]) for i in xrange(2 * N)])
al = tuple([(r[i], al[i], al[i]) for i in xrange(2 * N)])
cdict = {'red': rd, 'green': gr, 'blue': bl, 'alpha': al}
return colors.LinearSegmentedColormap('my_colormap', cdict)
def draw_colorbar(self,_axes,unique_vals=None,cax=None):
if unique_vals is None:
unique_vals = np.linspace(0, 1, 1000)
xmin, xmax = _axes.get_xlim()
ymin, ymax = _axes.get_ylim()
zmin, zmax = _axes.get_zlim()
invis=_axes.scatter(unique_vals,
unique_vals,
c=np.arange(len(unique_vals)),
cmap=self.colormap)
_axes.set_xlim([xmin,xmax])
_axes.set_ylim([ymin,ymax])
_axes.set_zlim([zmin,zmax])
cbar=self.fig.colorbar(invis,ax=_axes,cax=cax,drawedges=False)
cbar.set_ticks(np.linspace(0,np.size(unique_vals),5))
if unique_vals is not None:
cbar.set_ticklabels(np.around(np.linspace(0,np.max(unique_vals),5),2))
invis.set_alpha(0)
def create_opacity_colormap(self,principal_rgb_color, scale_size=256):
'''
Create opacity colormap based on one principal RGB color
'''
if np.any(principal_rgb_color > 1):
raise Exception('principal_rgb_color values should be in range [0,1]')
opac_colormap = np.concatenate((np.tile(principal_rgb_color[None, :],
(scale_size, 1))[:],
np.linspace(0, 1, scale_size)[:, None]),
axis=1)
self.colormap=self.array2cmap(opac_colormap)
def create_brightness_colormap(self,principal_rgb_color, scale_size):
'''
Create brightness colormap based on one principal RGB color
'''
if np.any(principal_rgb_color > 1):
raise Exception('principal_rgb_color values should be in range [0,1]')
hsv_color = colors.rgb_to_hsv(principal_rgb_color)
hsv_colormap = np.concatenate((np.tile(hsv_color[:-1][None, :], (scale_size, 1))[:],
np.linspace(0, 1, scale_size)[:, None]),
axis=1)
self.colormap=self.array2cmap(colors.hsv_to_rgb(hsv_colormap))
def draw(self,hist,edges,
fig=None,gs=None,subplot=None,
color=np.array([1,0,0]),all_axes=None):
'''
fig=figure handle
gs= contiguous slice (or whole) of gridspec to host plot
hist,edges=histogramdd output
'''
if fig is not None:
self.fig=fig
else:
self.fig=plt.figure()
if gs is None:
gs = gridspec.GridSpec(50, 50)
if all_axes is None:
_axes = self.fig.add_subplot(gs[:-5,:45],projection='3d')
cax=self.fig.add_subplot(gs[:-5,45:])
ax1=self.fig.add_subplot(gs[-4:-2,:])
ax2=self.fig.add_subplot(gs[-2:,:])
else:
_axes,cax,ax1,ax2=all_axes
_axes.clear()
cax.clear()
ax1.clear()
ax2.clear()
#unique_hist=np.unique(hist)
self.create_opacity_colormap(color)
self.draw_cubes(_axes, hist, edges)
self.draw_colorbar(_axes,cax=cax)
_axes.set_xlim((edges[0].min(),edges[0].max()))
_axes.set_ylim((edges[1].min(),edges[1].max()))
_axes.set_zlim((edges[2].min(),edges[2].max()))
self.fig.patch.set_facecolor('white')
_axes.w_xaxis.set_pane_color((0.8, 0.8, 0.8, 1.0))
_axes.w_yaxis.set_pane_color((0.8, 0.8, 0.8, 1.0))
_axes.w_zaxis.set_pane_color((0.8, 0.8, 0.8, 1.0))
if not self.save_only:
self.set_sliders(ax1, ax2)
return _axes,ax1,ax2
