Creating widgets with SpectAcoular¶

In this example, we are going to create an interactive plot of a Microphone Array Geometry with additional control widgets using the SpectAcoular package.

Note

For full interactivity, this example should be run as a Bokeh server application.

Set up the microphone geometry plot¶

Similar to the Basic plotting with Bokeh, we will first import the necessary modules and set up the microphone geometry plot.

import acoular as ac
import spectacoular as sp
from pathlib import Path
from bokeh.models import ColumnDataSource
from bokeh.plotting import figure

# set up the figure
figure = figure(title='Microphone Geometry', tools='hover,zoom_in,zoom_out,reset,lasso_select', match_aspect=True)

# set up the microphone geometry
default_xml_file = Path(ac.__file__).parent / 'xml' / 'tub_vogel64.xml'
mics = sp.MicGeom(file=default_xml_file)

# add circles to the figure to represent the microphones
cds = ColumnDataSource(data={'x': mics.pos_total[0], 'y': mics.pos_total[1]})
glyph = figure.circle(x='x', y='y', radius=0.02, line_color='black', fill_color='#1F77B4', fill_alpha=0.4, source=cds)

Create widgets for MicGeom¶

The MicGeom class provides several attributes, known as traits, that can be used to either provide additional information about the microphone array or to modify the geometry.

print(mics.traits().keys())

# Out:
# ['num_mics', 'center', 'aperture', 'file', 'pos_total', 'pos',  ...]

Let’s say we want to know about the number of microphones (num_mics) and the aperture of the microphone array (aperture), we can use SpectAcoular’s get_widgets() function to translate each attribute into a Bokeh widget.

Defining each trait to widget mapping individually can become cumbersome, and not all widgets are suitable for all traits. Therefore, SpectAcoular provides default widget mapping for most of Acoular’s classes.

from bokeh.models.widgets import NumericInput # noqa: E402

numeric_widgets = mics.get_widgets(
    {'aperture' : NumericInput, 'num_mics': NumericInput},
    {'aperture': {'title': 'Aperture/m', 'disabled': True},
     'num_mics': {'title': 'Number of Mics', 'disabled': True}})

It would also be beneficial to have a widget that allows the user to edit the microphone positions and to display the coodinates in a table format. Both can be achieved by using Bokeh’s DataTable widget. In addition, we will use the NumberEditor to allow the user to edit the values in the table.

from bokeh.models.widgets import DataTable # noqa: E402
from bokeh.models import TableColumn, NumberEditor # noqa: E402

editor = NumberEditor()
pos_table = [
    TableColumn(field='x', title='x/m', editor=editor),
    TableColumn(field='y', title='y/m', editor=editor),
    TableColumn(field='z', title='z/m', editor=editor)
        ]

# get the widgets for the MicGeom instance
data_table_widget = mics.get_widgets(
    {'pos_total': DataTable},
    {'pos_total': {'height': 450, 'transposed': True, 'columns': pos_table, 'source': cds, 'editable': True}}
    )

Finally, we want to allow the user to add or remove microphones from the microphone array. Therefore, we will use the PointDrawTool.

from bokeh.models import PointDrawTool # noqa: E402

draw_tool = PointDrawTool(renderers=[glyph], empty_value=0.)

figure.add_tools(draw_tool)
figure.toolbar.active_tap = draw_tool

Let’s create a layout that contains the figure and the widgets.

from bokeh.layouts import column, row # noqa: E402
from bokeh.io import show # noqa: E402

widget_column = column(*numeric_widgets.values(), *data_table_widget.values(), width=400)
layout = row(figure, widget_column)
show(layout)  # Show the plot in a new browser window

The following code is for Bokeh server apps only.

from bokeh.io import curdoc
curdoc().add_root(layout)  # Add the layout to the current document for Bokeh server apps

Gallery generated by Sphinx-Gallery

SpectAcoular documentation