def _theta_plot_default(self):
theta = self.theta
nclasses = theta.shape[0]
# create a plot data object and give it this data
plot_data = ArrayPlotData()
plot_data.set_data('classes', range(nclasses))
# create the plot
plot = Plot(plot_data)
# --- plot theta samples
if self.theta_samples is not None:
self._plot_samples(plot, plot_data)
# --- plot values of theta
plots = self._plot_theta_values(plot, plot_data)
# --- adjust plot appearance
plot.aspect_ratio = 1.6 if is_display_small() else 1.7
# adjust axis bounds
y_high = theta.max()
if self.theta_samples is not None:
y_high = max(y_high, self.theta_samples.max())
plot.range2d = DataRange2D(
low = (-0.2, 0.0),
high = (nclasses-1+0.2, y_high*1.1)
)
# create new horizontal axis
label_axis = self._create_increment_one_axis(
plot, 0., nclasses, 'bottom')
label_axis.title = 'True classes'
self._add_index_axis(plot, label_axis)
# label vertical axis
plot.value_axis.title = 'Probability'
# add legend
legend = Legend(component=plot, plots=plots,
align="ur", border_padding=10)
legend.tools.append(LegendTool(legend, drag_button="left"))
legend.padding_right = -100
plot.overlays.append(legend)
container = VPlotContainer(width=plot.width + 100, halign='left')
plot.padding_bottom = 50
plot.padding_top = 10
plot.padding_left = 0
container.add(plot)
container.bgcolor = 0xFFFFFF
self.decorate_plot(container, theta)
return container
def _theta_plot_default(self):
theta = self.theta
nclasses = theta.shape[0]
# create a plot data object and give it this data
plot_data = ArrayPlotData()
plot_data.set_data('classes', list(range(nclasses)))
# create the plot
plot = Plot(plot_data)
# --- plot theta samples
if self.theta_samples is not None:
self._plot_samples(plot, plot_data)
# --- plot values of theta
plots = self._plot_theta_values(plot, plot_data)
# --- adjust plot appearance
plot.aspect_ratio = 1.6 if is_display_small() else 1.7
# adjust axis bounds
y_high = theta.max()
if self.theta_samples is not None:
y_high = max(y_high, self.theta_samples.max())
plot.range2d = DataRange2D(low=(-0.2, 0.0),
high=(nclasses - 1 + 0.2, y_high * 1.1))
# create new horizontal axis
label_axis = self._create_increment_one_axis(plot, 0., nclasses,
'bottom')
label_axis.title = 'True classes'
self._add_index_axis(plot, label_axis)
# label vertical axis
plot.value_axis.title = 'Probability'
# add legend
legend = Legend(component=plot,
plots=plots,
align="ur",
border_padding=10)
legend.tools.append(LegendTool(legend, drag_button="left"))
legend.padding_right = -100
plot.overlays.append(legend)
container = VPlotContainer(width=plot.width + 100, halign='left')
plot.padding_bottom = 50
plot.padding_top = 10
plot.padding_left = 0
container.add(plot)
container.bgcolor = 0xFFFFFF
self.decorate_plot(container, theta)
return container
def _theta_plot_default(self):
"""Create plot of theta parameters."""
# We plot both the thetas and the samples from the posterior; if the
# latter are not defined, the corresponding ArrayPlotData names
# should be set to an empty list, so that they are not displayed
theta = self.model.theta
theta_len = theta.shape[0]
# create the plot data
if not self.theta_plot_data:
self.theta_plot_data = ArrayPlotData()
self._update_plot_data()
# create the plot
theta_plot = Plot(self.theta_plot_data)
for idx in range(theta_len):
# candle plot summarizing samples over the posterior
theta_plot.candle_plot((_w_idx('index', idx),
_w_idx('min', idx),
_w_idx('barmin', idx),
_w_idx('avg', idx),
_w_idx('barmax', idx),
_w_idx('max', idx)),
color = get_annotator_color(idx),
bar_line_color = "black",
stem_color = "blue",
center_color = "red",
center_width = 2)
# plot of raw samples
theta_plot.plot((_w_idx('ysamples', idx),
_w_idx('xsamples', idx)),
type='scatter',
color='black',
marker='dot',
line_width=0.5,
marker_size=1)
# plot current parameters
theta_plot.plot((_w_idx('y', idx), _w_idx('x', idx)),
type='scatter',
color=get_annotator_color(idx),
marker='plus',
marker_size=8,
line_width=2)
# adjust axis bounds
theta_plot.range2d = self._compute_range2d()
# remove horizontal grid and axis
theta_plot.underlays = [theta_plot.x_grid, theta_plot.y_axis]
# create new horizontal axis
label_list = [str(i) for i in range(theta_len)]
label_axis = LabelAxis(
theta_plot,
orientation = 'bottom',
positions = range(1, theta_len+1),
labels = label_list,
label_rotation = 0
)
# use a FixedScale tick generator with a resolution of 1
label_axis.tick_generator = ScalesTickGenerator(scale=FixedScale(1.))
theta_plot.index_axis = label_axis
theta_plot.underlays.append(label_axis)
theta_plot.padding = 25
theta_plot.padding_left = 40
theta_plot.aspect_ratio = 1.0
container = VPlotContainer()
container.add(theta_plot)
container.bgcolor = 0xFFFFFF
self.decorate_plot(container, theta)
self._set_title(theta_plot)
return container
def _theta_plot_default(self):
"""Create plot of theta parameters."""
# We plot both the thetas and the samples from the posterior; if the
# latter are not defined, the corresponding ArrayPlotData names
# should be set to an empty list, so that they are not displayed
theta = self.model.theta
theta_len = theta.shape[0]
# create the plot data
if not self.theta_plot_data:
self.theta_plot_data = ArrayPlotData()
self._update_plot_data()
# create the plot
theta_plot = Plot(self.theta_plot_data)
for idx in range(theta_len):
# candle plot summarizing samples over the posterior
theta_plot.candle_plot((_w_idx('index', idx),
_w_idx('min', idx),
_w_idx('barmin', idx),
_w_idx('avg', idx),
_w_idx('barmax', idx),
_w_idx('max', idx)),
color = get_annotator_color(idx),
bar_line_color = "black",
stem_color = "blue",
center_color = "red",
center_width = 2)
# plot of raw samples
theta_plot.plot((_w_idx('ysamples', idx),
_w_idx('xsamples', idx)),
type='scatter',
color='black',
marker='dot',
line_width=0.5,
marker_size=1)
# plot current parameters
theta_plot.plot((_w_idx('y', idx), _w_idx('x', idx)),
type='scatter',
color=get_annotator_color(idx),
marker='plus',
marker_size=8,
line_width=2)
# adjust axis bounds
theta_plot.range2d = self._compute_range2d()
# remove horizontal grid and axis
theta_plot.underlays = [theta_plot.x_grid, theta_plot.y_axis]
# create new horizontal axis
label_list = [str(i) for i in range(theta_len)]
label_axis = LabelAxis(
theta_plot,
orientation = 'bottom',
positions = list(range(1, theta_len+1)),
labels = label_list,
label_rotation = 0
)
# use a FixedScale tick generator with a resolution of 1
label_axis.tick_generator = ScalesTickGenerator(scale=FixedScale(1.))
theta_plot.index_axis = label_axis
theta_plot.underlays.append(label_axis)
theta_plot.padding = 25
theta_plot.padding_left = 40
theta_plot.aspect_ratio = 1.0
container = VPlotContainer()
container.add(theta_plot)
container.bgcolor = 0xFFFFFF
self.decorate_plot(container, theta)
self._set_title(theta_plot)
return container
def create_plot(self):
if hasattr(self.value, 'shadows'):
color_gen = color_generator()
shadowcolors = {}
for shadow in self.value.shadows:
shadowcolors[shadow] = color_gen.next()
container_class = {
'h': HPlotContainer,
'v': VPlotContainer
}[self.orientation]
container = container_class(spacing=15,
padding=15,
bgcolor='transparent')
container.fill_padding = True
container.bgcolor = (236 / 255.0, 233 / 255.0, 216 / 255.0)
if self.show_all:
self.plot_items = self.value.keys()
if len(self.plot_items) > 0:
plot_configs = []
for (plot_num, var_name) in enumerate(self.plot_items):
if not (isinstance(self.value[var_name], ndarray) and \
len(self.value[var_name].shape) == 1):
continue
plot_configs.append(
PlotConfig(x=var_name + '_index',
y=var_name,
type='Line',
number=plot_num))
self.plot_configs = plot_configs
if len(self.plot_configs) > 0:
number_to_plots = {}
for plot_config in self.plot_configs:
plotlist = number_to_plots.get(plot_config.number, [])
plotlist.append(plot_config)
number_to_plots[plot_config.number] = plotlist
keys = number_to_plots.keys()
keys.sort()
container_list = [number_to_plots[number] for number in keys]
for plot_group in container_list:
context_adapter = PlotDataContextAdapter(context=self.value)
plot = Plot(context_adapter)
plot.padding = 15
plot.padding_left = 35
plot.padding_bottom = 30
plot.spacing = 15
plot.border_visible = True
for plot_item in plot_group:
if len(self.value[plot_item.y].shape) == 2:
color_range = DataRange1D(
low=min(self.value[plot_item.y]),
high=max(self.value[plot_item.y]))
plot.img_plot(plot_item.y,
colormap=gray(color_range),
name=plot_item.y)
else:
plot_type = {
'Line': 'line',
'Scatter': 'scatter'
}[plot_item.type]
plot.plot(
(plot_item.x, plot_item.y),
name=plot_item.x + " , " + plot_item.y,
color=(.7, .7, .7),
type=plot_type,
)
if plot.index_axis.title != '':
plot.index_axis.title = plot.index_axis.title + ', ' + plot_item.x
else:
plot.index_axis.title = plot_item.x
if plot.value_axis.title != '':
plot.value_axis.title = plot.value_axis.title + ', ' + plot_item.y
else:
plot.value_axis.title = plot_item.y
if self.view_shadows and hasattr(
self.value, 'shadows'):
self.generate_shadow_plots(plot, shadowcolors,
plot_item, plot_type)
plot.tools.append(PanTool(plot))
container.add(plot)
self.plot = container
def create_plot(self):
if hasattr(self.value, 'shadows'):
color_gen = color_generator()
shadowcolors = {}
for shadow in self.value.shadows:
shadowcolors[shadow] = color_gen.next()
container_class = {'h' : HPlotContainer, 'v' : VPlotContainer}[self.orientation]
container = container_class(spacing=15, padding=15, bgcolor = 'transparent')
container.fill_padding = True
container.bgcolor=(236/255.0, 233/255.0, 216/255.0)
if self.show_all:
self.plot_items = self.value.keys()
if len(self.plot_items)>0:
plot_configs = []
for (plot_num, var_name) in enumerate(self.plot_items):
if not (isinstance(self.value[var_name], ndarray) and \
len(self.value[var_name].shape) == 1):
continue
plot_configs.append(PlotConfig(x=var_name + '_index',
y=var_name,
type='Line',
number=plot_num))
self.plot_configs = plot_configs
if len(self.plot_configs)>0:
number_to_plots = {}
for plot_config in self.plot_configs:
plotlist = number_to_plots.get(plot_config.number, [])
plotlist.append(plot_config)
number_to_plots[plot_config.number] = plotlist
keys = number_to_plots.keys()
keys.sort()
container_list = [number_to_plots[number] for number in keys]
for plot_group in container_list:
context_adapter = PlotDataContextAdapter(context=self.value)
plot = Plot(context_adapter)
plot.padding = 15
plot.padding_left=35
plot.padding_bottom = 30
plot.spacing=15
plot.border_visible = True
for plot_item in plot_group:
if len(self.value[plot_item.y].shape) == 2:
color_range = DataRange1D(low=min(self.value[plot_item.y]),
high=max(self.value[plot_item.y]))
plot.img_plot(plot_item.y, colormap=gray(color_range),
name=plot_item.y)
else:
plot_type = {'Line':'line', 'Scatter':'scatter'}[plot_item.type]
plot.plot((plot_item.x, plot_item.y),
name=plot_item.x + " , " + plot_item.y,
color=(.7, .7, .7),
type=plot_type,)
if plot.index_axis.title != '':
plot.index_axis.title = plot.index_axis.title + ', ' + plot_item.x
else:
plot.index_axis.title = plot_item.x
if plot.value_axis.title != '':
plot.value_axis.title = plot.value_axis.title + ', ' + plot_item.y
else:
plot.value_axis.title = plot_item.y
if self.view_shadows and hasattr(self.value, 'shadows'):
self.generate_shadow_plots(plot, shadowcolors, plot_item, plot_type)
plot.tools.append(PanTool(plot))
container.add(plot)
self.plot = container
