Source code for node_figure
# This file is part of Sympathy for Data.
# Copyright (c) 2016-2017, Combine Control Systems AB
#
# Sympathy for Data is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, version 3 of the License.
#
# Sympathy for Data is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Sympathy for Data. If not, see <http://www.gnu.org/licenses/>.
import numpy as np
from sympathy import api
from sympathy.api import node as synode
from sympathy.api import qt2 as qt_compat
from sympathy.api import node_helper
from sympathy.api.nodeconfig import Port, Ports, Tag, Tags, adjust
from sympathy.utils import preview
from sylib.figure import drawing, gui, mpl_utils
QtCore = qt_compat.QtCore
QtGui = qt_compat.QtGui
qt_compat.backend.use_matplotlib_qt()
[docs]class FigureCompressor(synode.Node):
"""
Compress a list of Figures into one Figure.
"""
author = 'Benedikt Ziegler'
version = '0.3'
icon = 'figurecompressor.svg'
name = 'Figure Compressor'
description = 'Compress a list of Figures to a single Figure'
nodeid = 'org.sysess.sympathy.visualize.figurecompressorgui'
tags = Tags(Tag.Visual.Figure)
related = ['org.sysess.sympathy.visualize.figuresubplot',
'org.sysess.sympathy.visualize.figure']
parameters = synode.parameters()
parameters.set_list(
'parent_figure', label='Parent figure:',
description='Specify the figure from which axes parameters '
'and legend position are copied.',
editor=synode.Util.combo_editor())
parameters.set_boolean(
'join_legends', value=True, label='Join legends',
description='Set if legends from different axes should be '
'joined into one legend.')
parameters.set_list(
'legend_location', value=[0], label='Legend position:',
plist=list(mpl_utils.LEGEND_LOC.keys()) + mpl_utils.OUTSIDE_LEGEND_LOC,
description='Defines the position of the joined legend.',
editor=synode.Util.combo_editor())
parameters.set_boolean(
'join_colorbars', value=False, label='Make first colorbar global',
description='If checked, the colorbar from the first figure becomes '
'a global colorbar in the output figure.')
parameters.set_boolean(
'auto_recolor', value=False, label='Auto recolor',
description='Automatically recolor all artists to avoid using a color '
'multiple times, if possible.')
parameters.set_boolean(
'auto_rescale', value=True, label='Auto rescale axes',
description='Automatically rescale all axes to fit the visible data.')
controllers = (
synode.controller(
when=synode.field('join_legends', 'checked'),
action=synode.field('legend_location', 'enabled')))
inputs = Ports([Port.Figures('List of Figures', name='input')])
outputs = Ports([Port.Figure(
'A Figure with the configured axes, lines, labels, etc',
name='figure')])
def adjust_parameters(self, node_context):
adjust(node_context.parameters['parent_figure'],
node_context.input['input'],
lists='index')
def execute(self, node_context):
input_figures = node_context.input['input']
output_figure = node_context.output['figure']
parameters = node_context.parameters
try:
parent_figure_number = parameters['parent_figure'].value[0]
except IndexError:
parent_figure_number = 0
input_axes = [figure.get_mpl_figure().axes for figure in input_figures]
default_output_axes = output_figure.first_subplot().get_mpl_axes()
axes_colorbars = drawing.compress_axes(
input_axes, default_output_axes,
parameters['join_legends'].value,
parameters['legend_location'].selected,
int(parent_figure_number),
auto_recolor=parameters['auto_recolor'].value,
auto_rescale=parameters['auto_rescale'].value,
add_colorbars=not parameters['join_colorbars'].value)
if parameters['join_colorbars'].value:
drawing.add_global_colorbar(axes_colorbars, output_figure)
[docs]class SubplotFigures(synode.Node):
"""
Layout the Figures in a list of Figures into subplots.
The number of rows and columns is automatically adjusted to an approximate
square. Empty axes in a non-empty row will be not shown.
"""
author = 'Benedikt Ziegler'
version = '0.2'
icon = 'figuresubplots.svg'
name = 'Layout Figures in Subplots'
description = 'Layout a list of Figures in a Subplot'
nodeid = 'org.sysess.sympathy.visualize.figuresubplot'
tags = Tags(Tag.Visual.Figure)
related = ['org.sysess.sympathy.visualize.figurecompressorgui',
'org.sysess.sympathy.visualize.figure']
inputs = Ports([Port.Figures('List of Figures', name='input')])
outputs = Ports([Port.Figure(
'A Figure with several subplot axes', name='figure')])
parameters = synode.parameters()
parameters.set_integer(
'rows', value=0, label='Number of rows',
description='Specify the number of rows, or 0 for auto.'
'If rows and columns are both 0, the node with attempt '
'to create an approximately square layout.',
editor=synode.Util.bounded_spinbox_editor(0, 100, 1))
parameters.set_integer(
'columns', value=0, label='Number of columns',
description='Specify the number of columns, or 0 for auto.'
'If rows and columns are both 0, the node with attempt '
'to create an approximately square layout.',
editor=synode.Util.bounded_spinbox_editor(0, 100, 1))
parameters.set_boolean(
'recolor', value=False, label='Auto recolor',
description='Specify if artists should be assigned new colors '
'automatically to prevent duplicate colors.')
# TODO(magnus): Using string here for compat with adjust(lists=index).
parameters.set_string(
'parent_figure', value='0', label='Parent figure:',
description='Specify the figure from which colorbar '
'and legend are copied.',
editor=synode.Util.combo_editor())
parameters.set_boolean(
'join_colorbars', value=False, label="Make parent's colorbar global",
description='If checked, the colorbar from the parent figure is '
'placed as a global colorbar for all subplots. '
'Note that it is up to you to make sure that the colorbar '
'is actually valid for all subplots.')
parameters.set_boolean(
'join_legends', value=False, label="Make parent's legend global",
description='If checked, the legend(s) in the first figure are '
'kept as global legends and all other legends are '
'discarded. Note that it is up to you to make sure that '
'the legend is actually valid for all subplots.')
parameters.set_string(
'share_x_axes', value='none', label='Share X axes',
description='If None, each subplot has an independent X axis. '
'Othewise, subplots will share X axis, meaning that '
'zooming/panning in one subplot also affects other '
'subplots.',
editor=synode.editors.combo_editor(
options={'none': 'None',
'col': 'Per column',
'all': 'All'}))
parameters.set_string(
'share_y_axes', value='none', label='Share Y axes',
description='If None, each subplot has an independent Y axis. '
'Othewise, subplots will share Y axis, meaning that '
'zooming/panning in one subplot also affects other '
'subplots.',
editor=synode.editors.combo_editor(
options={'none': 'None',
'row': 'Per row',
'all': 'All'}))
def update_parameters(self, params):
# Removed in 1.6.2:
if 'remove_internal_ticks' in params:
del params['remove_internal_ticks']
def adjust_parameters(self, node_context):
adjust(node_context.parameters['parent_figure'],
node_context.input['input'],
lists='index')
def execute(self, node_context):
input_figures = node_context.input['input']
output_figure = node_context.output['figure']
parameters = node_context.parameters
rows = parameters['rows'].value
cols = parameters['columns'].value
auto_recolor = parameters['recolor'].value
parent_figure = int(parameters['parent_figure'].value)
global_colorbar = parameters['join_colorbars'].value
global_legend = parameters['join_legends'].value
sharex = parameters['share_x_axes'].value
sharey = parameters['share_y_axes'].value
# calculate the number of rows and columns if any is =0
nb_input_figures = len(input_figures)
if rows == 0 and cols == 0:
rows = int(np.ceil(np.sqrt(nb_input_figures)))
cols = int(np.ceil(np.sqrt(nb_input_figures)))
if rows * cols - cols >= nb_input_figures > 0:
rows -= 1
elif rows == 0 and cols > 0:
rows = int(np.ceil(nb_input_figures / float(cols)))
elif rows > 0 and cols == 0:
cols = int(np.ceil(nb_input_figures / float(rows)))
if sharex == 'none' and sharey == 'none':
auto_rescale = False
elif sharex == 'none':
auto_rescale = 'y'
elif sharey == 'none':
auto_rescale = 'x'
else:
auto_rescale = True
subplots = np.array(output_figure.subplots(
rows, cols, sharex=sharex, sharey=sharey)).ravel()
figure_colorbars = {}
for i, (subplot, input_figure) in reversed(list(enumerate(
zip(subplots, input_figures)))):
default_axes = subplot.get_mpl_axes()
input_axes = [axes.get_mpl_axes() for axes in input_figure.axes]
axes_colorbars = drawing.compress_axes(
[input_axes], default_axes,
legends_join=False,
legend_location='best',
copy_properties_from=0,
auto_recolor=auto_recolor,
auto_rescale=auto_rescale,
add_colorbars=not global_colorbar,
add_legends=(i == parent_figure or not global_legend))
figure_colorbars[i] = axes_colorbars
if global_colorbar and parent_figure in figure_colorbars:
drawing.add_global_colorbar(
figure_colorbars[parent_figure], output_figure)
# don't show empty axes
if len(subplots) > len(input_figures):
for ax_to_blank in subplots[len(input_figures):]:
ax_to_blank.set_axis(False)
[docs]class Figure(synode.Node):
"""
The main configuration for this node consists of a tree view containing all
the parts of your figure. There are also simplified guis for creating
configurations for some common plot types. These are found in the wizard's
hat menu in the toolbar. Going back to the tree view from a simplified gui
lets you see the configuration that was created and modify it to suit your
needs. This is a great way to quickly create a plot or to learn how to do
some specific things in the tree view.
The tree view
=============
To add plots to the figure, you can either click on its corresponding
button in the toolbar, or press on a plot button and dragging it to where
in the tree view you want it (possible drop locations will be shown in
green). The plot will be added with some basic properties depending on
which plot type you added (e.g. *X Data* and *Y Data* for line plot).
Almost all configuration items support more than the default properties. To
add more, right-click on a configuration item and choose "Add...".
Some common plots
=================
There are many types of plots available in the Figure node. Here follows a
short word about some of the more common ones.
Scatter plots and line plots
----------------------------
These plot types are mostly pretty self-explanatory. The main difference
between these is while a *Line plot* can have marker, a *Scatter plot* can
not have lines. The *Scatter plot* can on the other hand have a different
size and color for each marker, whereas *Line plot* can only have a single
color and size for all markers.
Bar plots and histograms
------------------------
Bar plots and histograms are pretty similar plots. The difference lies in
how their x axis data is structured. The bar plot has distinct labels for
each bin, whereas each bin in a histogram lies between two points on a
continueous line. To get data on the correct format for a histogram plot
you can use the node :ref:`org.sysess.sympathy.dataanalysis.histogramcalc`
as a preprocessing node.
Use a *Bar Container* if you want to combine multiple bar plots by grouping
or stacking them. The bar plots should all have the same *Bar Labels*.
Please note that stacked bar plots are only situationally useful since it's
very difficult to gauge the heights of the individual bar parts.
Use a *Histogram Container* if you want to combine multiple histograms by
stacking them on top of each other. Please note that stacked histograms are
only situationally useful since it can be very difficult to read the
distributions of the individual histograms.
Heatmaps
--------
Heatmaps are two-dimensional histograms. To get data on the correct format
for a heatmap plot you can use the node
:ref:`org.sysess.sympathy.dataanalysis.heatmapcalc` as a preprocessing
node.
Box plots
---------
Box plots are a good for comparing different distributions of data. The box
plot is special in that it expects a *list* of arrays as data. It can for
example be specified as ``[arg['Column A'], arg['Column B']]`` or with a
list comprehension ``[arg[col_name] for col_name in arg.column_names()]``.
Pie charts
----------
Pie charts can be used to show parts of a whole, but are generally
considered inferior to e.g. bar plots. If you use a pie chart you will also
want to set the *Aspect ratio* of the *Axes* to 'equal'. Otherwise your pie
chart will be very hard to ready accurately.
Multiple axes
=============
There is support in the node for having more than one pair of axes. This
can for example be useful if you want to have two different y scales for
two line plots. To get two y axes first add an extra *Axes* item to the
configuration tree. In the new *Axes*, you should set the *position* of the
*YAxis* to 'right'. Lastly add one line plot to each *Axes*.
Legends
=======
To get a legend you must both set labels on all plots that you want to
include in the legend and also add a *Legend* item. Note that some plot
types have other configuration items called things like *Bin Labels* or
*Bar Labels*, but those are not used for the legend. Instead, look for a
property called simply *Label* or *Labels*.
It is possible to place the legend outside of the axes, but you might need
to tweak the *Distance from Axes* property to get it to look just right for
your specific plot.
To get a single legend which summarizes all signals across multiple *Axes*
you need to add the *Legend* to the root of the configuration tree, i.e.
next to the *Axes* items.
Iterators
=========
An *Iterator* can be used to create a dynamic number of plots in a single
figure. The iterator can only contain a single plot, but that plot will
then be repeated depending on the *Iterable*. For example if the data is a
Table and the *Iterable* is ``col_name = arg.column_names()`` then
``col_name`` will take on all the column_names in the Table and can be used
in e.g. *Y Data* as such: ``arg[col_name]``.
Python expressions as values
============================
To allow the user extra flexibility, many properties can be given either as
a normal value or as a python expression which evaluates to a value for
that configuration item. To swith to Python mode click the grayed out
Python icon to the right of the normal input field. Note that some fields
can only be entered as Python expressions (notably fields like *X Data* and
*Y Data* for line plots).
In the python evironment the input data port is available under the name
``arg``. For example one can refer to data columns in a connected Table by
writing something like ``arg['My data column']``. Have a look at the
:ref:`datatypeapis` to see all the available methods and attributes for the
data type that you connect to the node.
Related nodes
=============
This node can not create subplots, but by creating multiple figure objects
you can use the node :ref:`org.sysess.sympathy.visualize.figuresubplot` to
arrange them as subplots.
Use the node :ref:`Export figures` to write the figures you produce to
files.
:ref:`org.sysess.sympathy.visualize.figurecompressorgui` can be used to
combine multiple figures into a single figure.
Known issues
============
Using Timeline Plot with datetime for X Data requires matplotlib
version >= 3.1.0, and will otherwise result in an exception.
"""
author = 'Benedikt Ziegler & Magnus Sandén'
version = '0.1'
icon = 'figure.svg'
name = 'Figure'
description = 'Create a Figure from some data.'
nodeid = 'org.sysess.sympathy.visualize.figure'
tags = Tags(Tag.Visual.Figure)
related = ['org.sysess.sympathy.visualize.figures',
'org.sysess.sympathy.visualize.figuresubplot',
'org.sysess.sympathy.visualize.figurecompressorgui']
parameters = synode.parameters()
parameters.set_json(
'parameters', value={},
description='The full configuration for this figure.')
inputs = Ports([Port.Custom('<a>', 'Input', name='input')])
outputs = Ports([
Port.Figure('Output figure', name='figure', preview=True)])
def _parameter_view(self, node_context, input_data):
figure_widget = gui.FigureFromTableWidget(
input_data, node_context.parameters['parameters'])
preview_widget = preview.PreviewWidget(
self, node_context, node_context.parameters)
widget = preview.ParameterPreviewWidget(
figure_widget, preview_widget)
return widget
def exec_parameter_view(self, node_context):
input_data = node_context.input['input']
if not input_data.is_valid():
input_data = api.table.File()
return self._parameter_view(node_context, input_data)
def execute(self, node_context):
data_table = node_context.input['input']
figure = node_context.output['figure']
config_table = node_context.output.group('config')
if len(config_table) > 0:
config_table = config_table[0]
else:
config_table = None
parameters = node_context.parameters['parameters'].value
figure_creator = drawing.CreateFigure(data_table, figure, parameters)
figure_creator.create_figure()
[docs]@node_helper.list_node_decorator(['input'], ['figure'])
class Figures(Figure):
name = 'Figures'
nodeid = 'org.sysess.sympathy.visualize.figures'
def _parameter_view_no_preview(self, node_context, input_data):
figure_widget = gui.FigureFromTableWidget(
input_data, node_context.parameters['parameters'])
preview_widget = preview.NullPreviewWidget(
'<b>Preview requires non-empty input list</b>')
preview_widget.setWordWrap(True)
preview_widget.setAlignment(
QtCore.Qt.AlignTop)
widget = preview.ParameterPreviewWidget(
figure_widget, preview_widget)
return widget
def exec_parameter_view(self, node_context):
input_data = node_context.input['input']
if input_data.is_valid() and len(input_data):
return self._parameter_view(node_context, input_data[0])
else:
input_data = api.table.File()
return self._parameter_view_no_preview(node_context, input_data)
