def fft_worker(self):
if _fft_worker_cache[0] != self.viewer.data.shape:
timeit = TimeIt(self.viewer, 'creating FFT worker')
_fft_worker_cache[0] = self.viewer.data.shape
_fft_worker_cache[1] = FFTTasks(self.viewer.data.shape,
options=Options(fftw_threads=4))
timeit.stop()
return _fft_worker_cache[1]
def runner (parser, options, args):
options = Options (options)
if not hasattr(parser, 'runner'):
options.output_path = None
if args:
if len (args)==1:
if options.input_path:
print >> sys.stderr, "WARNING: overwriting input path %r with %r" % (options.input_path, args[0])
options.input_path = args[0]
if options.input_path is None:
parser.error('Expected --input-path but got nothing')
options.input_path = fix_path (options.input_path)
stack = ImageStack.load(options.input_path, options=options)
images = stack.images
roll_axis = dict(XY=0, XZ=1, YZ=2).get(options.projection, 0)
axis_labels = ('Z', 'Y', 'X')
#roll_axis = int (options.roll_axis)
voxel_sizes = stack.get_voxel_sizes()
resolution = ['', '', '']
title = '%s[%s]' % (os.path.basename (options.input_path), images.dtype)
dr = stack.get_lateral_resolution()
dz = stack.get_axial_resolution()
if voxel_sizes:
if dr is not None:
resolution[1] = tostr(dr/voxel_sizes[1]) + 'px'
resolution[2] = tostr(dr/voxel_sizes[2]) + 'px'
if dz is not None:
resolution[0] = tostr(dz/voxel_sizes[0]) + 'px'
resolution = tuple (resolution)
if roll_axis:
images = numpy.rollaxis(images, roll_axis)
if voxel_sizes:
voxel_sizes = (voxel_sizes[roll_axis],) + voxel_sizes[:roll_axis] + voxel_sizes[roll_axis+1:]
axis_labels = (axis_labels[roll_axis],) + axis_labels[:roll_axis] + axis_labels[roll_axis+1:]
resolutions = (resolution[roll_axis],) + resolution[:roll_axis] + resolution[roll_axis+1:]
if voxel_sizes:
xlabel = '%s, resol=%s, px size=%sum, size=%sum' \
% (axis_labels[-1], resolution[-1], tostr(voxel_sizes[-1]*1e6), tostr(voxel_sizes[-1]*1e6*images.shape[-1]))
ylabel = '%s, resol=%s, px size=%sum, size=%sum' \
% (axis_labels[-2], resolution[-2], tostr(voxel_sizes[-2]*1e6), tostr(voxel_sizes[-2]*1e6*images.shape[-2]))
else:
xlabel = '%s' % (axis_labels[-1])
ylabel = '%s' % (axis_labels[-2])
import matplotlib.cm as cm
import matplotlib.pyplot as pyplot
from iocbio.io.tifffile import imshow
if options.invert_cmap:
cmap = getattr(cm, options.cmap+'_r', options.cmap)
else:
cmap = getattr(cm, options.cmap, 'gray')
view_3d = options.get(view_3d = 'none')
if view_3d and view_3d.lower()=='none':
view_3d = None
if view_3d:
l = []
for i,d in enumerate(view_3d.split (',')):
d = d.strip()
if d.lower()=='c':
d = images.shape[i]//2
else:
try:
d = int(d)
except:
d = images.shape[i]//2
d = max(min(images.shape[i]-1, d), 0)
l.append(d)
view_3d = l
def mark_image (image, c1, c2, ln, wd, mx):
image[c1-wd:c1+wd+1,:ln] = mx
image[c1-wd:c1+wd+1,-ln:] = mx
image[:ln,c2-wd:c2+wd+1] = mx
image[-ln:,c2-wd:c2+wd+1] = mx
return image
if view_3d:
import scipy.ndimage as ndimage
pyplot.rc('font', family='sans-serif', weight='normal', size=8)
figure = pyplot.figure(dpi=options.get(dpi=96),
figsize=(8, 8), frameon=True,
facecolor='1.0', edgecolor='w')
yz_scale = voxel_sizes[1] / voxel_sizes[0]
zx_scale = voxel_sizes[0] / voxel_sizes[2]
yx_scale = voxel_sizes[1] / voxel_sizes[2]
yx_image = images[view_3d[0]].copy ()
zx_image = images[:,view_3d[1],:].copy ()
yz_image = images[:,:,view_3d[2]].T.copy()
image = numpy.zeros((yx_image.shape[0] + zx_image.shape[0]+1, yx_image.shape[1]+yz_image.shape[1]+1))
wd = image.shape[0]//300
ln = max(1, image.shape[1]//20)
mx = yx_image.max()
def fix_image (image, scale, c1, c2, ln, wd):
mx = image.max()
if scale==1:
mark_image (image, c1, c2, ln, wd, mx)
elif scale>1:
image = ndimage.interpolation.zoom(image, [scale, 1.0], order=0)
mark_image (image, c1*scale, c2, ln, wd, mx)
else:
image = ndimage.interpolation.zoom(image, [1.0, 1.0/scale], order=0)
mark_image (image, c1, c2/scale, ln, wd, mx)
return image
yx_image = fix_image(yx_image, yx_scale, view_3d[1], view_3d[2], ln, wd)
zx_image = fix_image(zx_image, zx_scale, view_3d[0], view_3d[2], ln, wd)
yz_image = fix_image(yz_image, yz_scale, view_3d[1], view_3d[0], ln, wd)
image = numpy.zeros((yx_image.shape[0] + zx_image.shape[0]+1, yx_image.shape[1]+yz_image.shape[1]+1))
image[:yx_image.shape[0], :yx_image.shape[1]] = yx_image
image[:yx_image.shape[0], yx_image.shape[1]+1:] = yz_image
image[yx_image.shape[0]+1:, :yx_image.shape[1]] = zx_image
image_plot = pyplot.imshow(image,
interpolation=options.get(interpolation='nearest'),
cmap = cmap,
)
pyplot.title(title, size=11)
axes = pyplot.gca()
xtickdata = [(0,'0'),
#(yx_image.shape[1]/2, 'X'),
(yx_image.shape[1], '%.1fum' % (voxel_sizes[2]*images.shape[2]*1e6)),
( (yx_image.shape[1]+image.shape[1])/2, 'Z'),
(image.shape[1]-1, '%.1fum' % (voxel_sizes[0]*images.shape[0]*1e6)),
(yx_image.shape[1]*view_3d[2]/images.shape[2],'X=%spx' % (view_3d[2])),
]
ytickdata = [(0,'0'),
#(yx_image.shape[0]/2, 'Y'),
(yx_image.shape[0], '%.1fum' % (voxel_sizes[1]*images.shape[1]*1e6)),
#((yx_image.shape[0]+image.shape[0])/2, 'Z'),
( image.shape[0]-1, '%.1fum' % (voxel_sizes[0]*images.shape[0]*1e6)),
(yx_image.shape[0]*view_3d[1]/images.shape[1],'Y=%spx' % (view_3d[1])),
(yx_image.shape[0]+yz_image.shape[1]*view_3d[0]/images.shape[0], 'Z=%spx' % (view_3d[0])),
]
xtickdata.sort ()
xticks, xticklabels = zip(*xtickdata)
ytickdata.sort ()
yticks, yticklabels = zip(*ytickdata)
axes.set_xticks(xticks)
axes.set_xticklabels (xticklabels)
axes.set_yticks(yticks)
axes.set_yticklabels (yticklabels)
cbar = pyplot.colorbar(shrink=0.8)
def on_keypressed(event):
"""Callback for key press event."""
key = event.key
if key == 'q':
sys.exit(0)
figure.canvas.mpl_connect('key_press_event', on_keypressed)
else:
figure, subplot, image = imshow(images, title = title,
#miniswhite=page.photometric=='miniswhite',
interpolation=options.interpolation,
cmap=cmap,
dpi=options.dpi, isrgb=options.rgb,
show_hist = options.histogram_bins,
auto_scale = options.auto_scale)
axes = figure.get_axes()[0]
axes.set_xlabel(xlabel)
axes.set_ylabel(ylabel)
output_path = options.get (output_path='')
if output_path:
pyplot.savefig (output_path)
print 'wrote',output_path
sys.exit(0)
pyplot.show()
def runner(parser, options, args):
options = Options(options)
if not hasattr(parser, 'runner'):
options.output_path = None
if args:
if len(args) == 1:
if options.input_path:
print >> sys.stderr, "WARNING: overwriting input path %r with %r" % (
options.input_path, args[0])
options.input_path = args[0]
if options.input_path is None:
parser.error('Expected --input-path but got nothing')
options.input_path = fix_path(options.input_path)
stack = ImageStack.load(options.input_path, options=options)
images = stack.images
roll_axis = dict(XY=0, XZ=1, YZ=2).get(options.projection, 0)
axis_labels = ('Z', 'Y', 'X')
#roll_axis = int (options.roll_axis)
voxel_sizes = stack.get_voxel_sizes()
resolution = ['', '', '']
title = '%s[%s]' % (os.path.basename(options.input_path), images.dtype)
dr = stack.get_lateral_resolution()
dz = stack.get_axial_resolution()
if voxel_sizes:
if dr is not None:
resolution[1] = tostr(dr / voxel_sizes[1]) + 'px'
resolution[2] = tostr(dr / voxel_sizes[2]) + 'px'
if dz is not None:
resolution[0] = tostr(dz / voxel_sizes[0]) + 'px'
resolution = tuple(resolution)
if roll_axis:
images = numpy.rollaxis(images, roll_axis)
if voxel_sizes:
voxel_sizes = (
voxel_sizes[roll_axis],
) + voxel_sizes[:roll_axis] + voxel_sizes[roll_axis + 1:]
axis_labels = (axis_labels[roll_axis],
) + axis_labels[:roll_axis] + axis_labels[roll_axis +
1:]
resolutions = (resolution[roll_axis],
) + resolution[:roll_axis] + resolution[roll_axis + 1:]
if voxel_sizes:
xlabel = '%s, resol=%s, px size=%sum, size=%sum' \
% (axis_labels[-1], resolution[-1], tostr(voxel_sizes[-1]*1e6), tostr(voxel_sizes[-1]*1e6*images.shape[-1]))
ylabel = '%s, resol=%s, px size=%sum, size=%sum' \
% (axis_labels[-2], resolution[-2], tostr(voxel_sizes[-2]*1e6), tostr(voxel_sizes[-2]*1e6*images.shape[-2]))
else:
xlabel = '%s' % (axis_labels[-1])
ylabel = '%s' % (axis_labels[-2])
import matplotlib.cm as cm
import matplotlib.pyplot as pyplot
from iocbio.io.tifffile import imshow
if options.invert_cmap:
cmap = getattr(cm, options.cmap + '_r', options.cmap)
else:
cmap = getattr(cm, options.cmap, 'gray')
view_3d = options.get(view_3d='none')
if view_3d and view_3d.lower() == 'none':
view_3d = None
if view_3d:
l = []
for i, d in enumerate(view_3d.split(',')):
d = d.strip()
if d.lower() == 'c':
d = images.shape[i] // 2
else:
try:
d = int(d)
except:
d = images.shape[i] // 2
d = max(min(images.shape[i] - 1, d), 0)
l.append(d)
view_3d = l
def mark_image(image, c1, c2, ln, wd, mx):
image[c1 - wd:c1 + wd + 1, :ln] = mx
image[c1 - wd:c1 + wd + 1, -ln:] = mx
image[:ln, c2 - wd:c2 + wd + 1] = mx
image[-ln:, c2 - wd:c2 + wd + 1] = mx
return image
if view_3d:
import scipy.ndimage as ndimage
pyplot.rc('font', family='sans-serif', weight='normal', size=8)
figure = pyplot.figure(dpi=options.get(dpi=96),
figsize=(8, 8),
frameon=True,
facecolor='1.0',
edgecolor='w')
yz_scale = voxel_sizes[1] / voxel_sizes[0]
zx_scale = voxel_sizes[0] / voxel_sizes[2]
yx_scale = voxel_sizes[1] / voxel_sizes[2]
yx_image = images[view_3d[0]].copy()
zx_image = images[:, view_3d[1], :].copy()
yz_image = images[:, :, view_3d[2]].T.copy()
image = numpy.zeros((yx_image.shape[0] + zx_image.shape[0] + 1,
yx_image.shape[1] + yz_image.shape[1] + 1))
wd = image.shape[0] // 300
ln = max(1, image.shape[1] // 20)
mx = yx_image.max()
def fix_image(image, scale, c1, c2, ln, wd):
mx = image.max()
if scale == 1:
mark_image(image, c1, c2, ln, wd, mx)
elif scale > 1:
image = ndimage.interpolation.zoom(image, [scale, 1.0],
order=0)
mark_image(image, c1 * scale, c2, ln, wd, mx)
else:
image = ndimage.interpolation.zoom(image, [1.0, 1.0 / scale],
order=0)
mark_image(image, c1, c2 / scale, ln, wd, mx)
return image
yx_image = fix_image(yx_image, yx_scale, view_3d[1], view_3d[2], ln,
wd)
zx_image = fix_image(zx_image, zx_scale, view_3d[0], view_3d[2], ln,
wd)
yz_image = fix_image(yz_image, yz_scale, view_3d[1], view_3d[0], ln,
wd)
image = numpy.zeros((yx_image.shape[0] + zx_image.shape[0] + 1,
yx_image.shape[1] + yz_image.shape[1] + 1))
image[:yx_image.shape[0], :yx_image.shape[1]] = yx_image
image[:yx_image.shape[0], yx_image.shape[1] + 1:] = yz_image
image[yx_image.shape[0] + 1:, :yx_image.shape[1]] = zx_image
image_plot = pyplot.imshow(
image,
interpolation=options.get(interpolation='nearest'),
cmap=cmap,
)
pyplot.title(title, size=11)
axes = pyplot.gca()
xtickdata = [
(0, '0'),
#(yx_image.shape[1]/2, 'X'),
(yx_image.shape[1],
'%.1fum' % (voxel_sizes[2] * images.shape[2] * 1e6)),
((yx_image.shape[1] + image.shape[1]) / 2, 'Z'),
(image.shape[1] - 1,
'%.1fum' % (voxel_sizes[0] * images.shape[0] * 1e6)),
(yx_image.shape[1] * view_3d[2] / images.shape[2],
'X=%spx' % (view_3d[2])),
]
ytickdata = [
(0, '0'),
#(yx_image.shape[0]/2, 'Y'),
(yx_image.shape[0],
'%.1fum' % (voxel_sizes[1] * images.shape[1] * 1e6)),
#((yx_image.shape[0]+image.shape[0])/2, 'Z'),
(image.shape[0] - 1,
'%.1fum' % (voxel_sizes[0] * images.shape[0] * 1e6)),
(yx_image.shape[0] * view_3d[1] / images.shape[1],
'Y=%spx' % (view_3d[1])),
(yx_image.shape[0] +
yz_image.shape[1] * view_3d[0] / images.shape[0],
'Z=%spx' % (view_3d[0])),
]
xtickdata.sort()
xticks, xticklabels = zip(*xtickdata)
ytickdata.sort()
yticks, yticklabels = zip(*ytickdata)
axes.set_xticks(xticks)
axes.set_xticklabels(xticklabels)
axes.set_yticks(yticks)
axes.set_yticklabels(yticklabels)
cbar = pyplot.colorbar(shrink=0.8)
def on_keypressed(event):
"""Callback for key press event."""
key = event.key
if key == 'q':
sys.exit(0)
figure.canvas.mpl_connect('key_press_event', on_keypressed)
else:
figure, subplot, image = imshow(
images,
title=title,
#miniswhite=page.photometric=='miniswhite',
interpolation=options.interpolation,
cmap=cmap,
dpi=options.dpi,
isrgb=options.rgb,
show_hist=options.histogram_bins,
auto_scale=options.auto_scale)
axes = figure.get_axes()[0]
axes.set_xlabel(xlabel)
axes.set_ylabel(ylabel)
output_path = options.get(output_path='')
if output_path:
pyplot.savefig(output_path)
print 'wrote', output_path
sys.exit(0)
pyplot.show()
def runner (parser, options, args):
options = Options (options)
run_method = getattr(parser, 'run_method', 'subcommand')
if os.name=='posix' and run_method == 'subprocess':
print 'This script cannot be run using subprocess method. Choose subcommand to continue.'
return
if args:
if len (args)==1:
if options.input_path:
print >> sys.stderr, "WARNING: overwriting input path %r with %r" % (options.input_path, args[0])
options.input_path = args[0]
if options.input_path is None:
parser.error('Expected --input-path but got nothing')
data, titles = RowFile (options.input_path).read(with_titles = True)
if options.print_keys:
print 'rowfile keys:' + ', '.join(data.keys())
print 'length:', len(data[data.keys()[0]] or [])
return
if not options.x_keys and not options.y_keys:
options.x_keys = titles[0]
options.y_keys = ','.join(titles[1:])
parser.save_options(options, [])
if not options.x_keys or not options.y_keys:
print 'No x or y keys specified for the plot.'
return
x_keys = [str (k).strip() for k in options.x_keys.split (',')]
y_keys = [str (k).strip() for k in options.y_keys.split (',')]
from mpl_toolkits.axes_grid.parasite_axes import HostAxes, ParasiteAxes
import matplotlib.pyplot as plt
fig = plt.figure(1, figsize=(12,6))
host = HostAxes(fig, [0.1, 0.1, 0.55, 0.8])
host.axis["right"].set_visible(False)
fig.add_axes(host)
host.set_title(options.input_path)
legend_list = []
axes_ylim = []
plot_axes = []
for x_key in x_keys:
x_data = data.get(x_key)
if x_data is None:
print 'Rowfile does not contain a column named %r' % (x_key)
continue
offset = 20
host.set_xlabel(x_key)
for y_key in y_keys:
plot_str = 'plot'
if y_key.startswith('log:'):
y_key = y_key[4:]
plot_str = 'semilogy'
y_data = data.get(y_key)
if y_data is None:
print 'Rowfile does not contain a column named %r' % (y_key)
continue
if axes_ylim:
ax = ParasiteAxes(host, sharex=host)
host.parasites.append(ax)
#ax.axis["right"].set_visible(True)
#ax.axis["right"].major_ticklabels.set_visible(True)
#ax.axis["right"].label.set_visible(True)
new_axisline = ax._grid_helper.new_fixed_axis
ax_line = ax.axis["right2"] = new_axisline(loc="right",
axes=ax,
offset=(offset,0))
if plot_str=='semilogy':
offset += 60
else:
offset += 40
else:
ax = host
ax_line = ax.axis['left']
ax.set_ylabel(y_key)
p, = getattr(ax, plot_str)(x_data, y_data, label = y_key)
axes_ylim.append((ax, min (y_data), max (y_data)))
plot_axes.append((p, ax_line))
[ax.set_ylim(mn,mx) for ax,mn,mx in axes_ylim]
#[ax_line.label.set_color (p.get_color()) for p, ax in plot_axes]
host.legend()
def on_keypressed(event):
key = event.key
if key=='q':
sys.exit(0)
fig.canvas.mpl_connect('key_press_event', on_keypressed)
plt.draw()
output_path = options.get (output_path='')
if output_path:
plt.savefig (output_path)
print 'wrote',output_path
sys.exit(0)
plt.show()