Insets and panels

Panel axes

It is often useful to have narrow “panels” along the edge of a larger subplot for plotting secondary 1-dimensional datasets or summary statistics. In matplotlib, there is no simple way to do this. In ProPlot, you can create panels by passing a location (e.g. loc='r' or loc='right') to the panel or panel_axes methods. The resulting axes are instances of CartesianAxes.

To generate “stacked” panels, simply call panel more than once. To include panels when centering spanning axis labels and super titles, pass includepanels=True to Figure. Panels do not interfere with the tight layout algorithm and do not affect the subplot aspect ratios.

In the first example below, the panel distance from the main subplot is manually set to space=0. In the second example, it is adjusted automatically by the tight layout algorithm.

import proplot as plot
import numpy as np
state = np.random.RandomState(51423)
data = (state.rand(20, 20) - 0.48).cumsum(axis=1).cumsum(axis=0)
data = 10 * (data - data.min()) / (data.max() - data.min())

# Stacked panels with outer colorbars
for cbarlocation, plocation in ('rb', 'br'):
    fig, axs = plot.subplots(
        axwidth=1.6, nrows=1, ncols=2,
        share=0, panelpad=0.1, includepanels=True
    )
    axs.contourf(
        data, cmap='glacial', extend='both',
        colorbar=cbarlocation, colorbar_kw={'label': 'colorbar'},
    )

    # Summary statistics and settings
    titleloc = 'upper center'
    axis = int(plocation == 'r')  # dimension along which stats are taken
    x1 = x2 = np.arange(20)
    y1 = data.mean(axis=axis)
    y2 = data.std(axis=axis)
    if plocation == 'r':
        titleloc = 'center'
        x1, x2, y1, y2 = y1, y2, x1, x2
    kwargs = {'titleloc': titleloc, 'xreverse': False, 'yreverse': False}
    space = 0
    width = '30pt'

    # Panels for plotting the mean
    panels = axs.panel(plocation, space=space, width=width)
    panels.plot(x1, y1, color='gray7')
    panels.format(title='Mean', **kwargs)

    # Panels for plotting the standard deviation
    panels = axs.panel(plocation, space=space, width=width)
    panels.plot(x2, y2, color='gray7', ls='--')
    panels.format(title='Stdev', **kwargs)

    # Apply formatting *after*
    axs.format(
        xlabel='xlabel', ylabel='ylabel', title='Title',
        suptitle='Using panels for summary statistics',
    )

import proplot as plot
fig, axs = plot.subplots(axwidth=1.5, nrows=2, ncols=2, share=0)

# Demonstration that complex arrangements of panels
# do not mess up tight layout algorithm
for ax, side in zip(axs, 'tlbr'):
    ax.panel(side, width='3em')
axs.format(
    xlim=(0, 1), ylim=(0, 1),
    xlabel='xlabel', ylabel='ylabel',
    yticks=plot.arange(0.2, 0.8, 0.2),
    xticks=plot.arange(0.2, 0.8, 0.2),
    title='Title', suptitle='Complex arrangement of panels',
    collabels=['Column 1', 'Column 2'],
    abc=True, abcloc='ul', titleloc='uc', abovetop=False,
)

Inset axes

Inset axes can be generated with the inset or inset_axes command. By defaut, the resulting axes use the same projection as the parent axes, but you can also specify a different projection (for example, ax.inset(bounds, proj='polar')). Passing zoom=True to inset draws “zoom indication” lines with indicate_inset_zoom when the axes are both Cartesian, and ProPlot automatically updates the lines when the axis limits of the parent axes change. To modify the line properties, simply use the zoom_kw argument.

import proplot as plot
import numpy as np

# Generate sample data
N = 20
state = np.random.RandomState(51423)
x, y = np.arange(10), np.arange(10)
data = state.rand(10, 10)

# Plot sample data
fig, ax = plot.subplots(axwidth=3)
m = ax.pcolormesh(data, cmap='Grays', levels=N)
ax.colorbar(m, loc='b', label='label')
ax.format(
    xlabel='xlabel', ylabel='ylabel',
    suptitle='"Zooming in" with an inset axes'
)

# Create inset axes representing a "zoom-in"
iax = ax.inset(
    [5, 5, 4, 4], transform='data', zoom=True,
    zoom_kw={'color': 'red3', 'lw': 2, 'ls': '--'}
)
iax.format(
    xlim=(2, 4), ylim=(2, 4), color='red7',
    linewidth=1.5, ticklabelweight='bold'
)
iax.pcolormesh(data, cmap='Grays', levels=N)

<matplotlib.collections.QuadMesh at 0x7f3191858910>