Colorbars and legends

    In matplotlib, colorbars are added to the edges of subplots with the figure method using e.g. fig.colorbar(m, ax=ax, location='right'). In ProPlot, this is done using the new axes colorbar method with e.g. ax.colorbar(m, loc='r'). The proplot.axes.Axes.colorbar method preserves subplot aspect ratios and visual symmetry between subplots by allocating new space in the figure rather than “stealing” space from the parent subplot (see the section on automatic subplot spacing for details).

    ProPlot tries to make the usage of consistent with legend, and includes an improved legend method that tries to do the same:

    • Just like colorbar, can draw “outer” legends along the edges of subplots when you request a side location for the legend (e.g. loc='right' or loc='r'). If you draw multiple colorbars and legends on one side, they are “stacked” on top of each other.

    • Just like , proplot.axes.Axes.colorbar can draw “inset” colorbars when you request an for the colorbar (e.g. loc='upper right' or loc='ur'). Inset colorbars have optional background “frames” that can be configured with various colorbar keywords.

    You can also draw colorbars and legends on-the-fly by supplying keyword arguments to various plotting commands. To plot data and draw a colorbar in one go, pass a location (e.g. colorbar='r') to methods that accept a cmap argument (e.g. ). To draw a legend or colorbar-legend in one go, pass a location (e.g. legend='r' or colorbar='r') to methods that accept a cycle argument (e.g. plot). Use legend_kw and colorbar_kw to pass keyword arguments to the colorbar and legend functions. This feature is powered by the and cycle_changer wrappers.

    1. import proplot as plot
    2. import numpy as np
    3. with plot.rc.context(abc=True):
    4.     fig, axs = plot.subplots(ncols=2, share=0)
    6. # Colorbars
    7. ax = axs[0]
    8. state = np.random.RandomState(51423)
    9. m = ax.heatmap(state.rand(10, 10), colorbar='t', cmap='dusk')
    10. ax.colorbar(m, loc='r')
    11. ax.colorbar(m, loc='ll', label='colorbar label')
    12. ax.format(title='Axes colorbars', suptitle='Axes colorbars and legends demo')
    14. # Legends
    15. ax = axs[1]
    16. ax.format(title='Axes legends', titlepad='0em')
    17. hs = ax.plot(
    18.     (state.rand(10, 5) - 0.5).cumsum(axis=0), linewidth=3,
    19.     cycle='ggplot', legend='t',
    20.     labels=list('abcde'), legend_kw={'ncols': 5, 'frame': False}
    21. )
    22. ax.legend(hs, loc='r', ncols=1, frame=False)
    23. ax.legend(hs, loc='ll', label='legend label')
    24. axs.format(xlabel='xlabel', ylabel='ylabel')
    1. [2]:
    1. import proplot as plot
    2. import numpy as np
    3. fig, axs = plot.subplots(nrows=2, share=0, axwidth='55mm', panelpad='1em')
    4. axs.format(suptitle='Stacked colorbars demo')
    5. state = np.random.RandomState(51423)
    6. N = 10
    7. # Repeat for both axes
    8. for j, ax in enumerate(axs):
    9.     ax.format(
    10.         xlabel='data', xlocator=np.linspace(0, 0.8, 5),
    11.         title=f'Subplot #{j+1}'
    12.     )
    13.     for i, (x0, y0, x1, y1, cmap, scale) in enumerate((
    14.         (0, 0.5, 1, 1, 'grays', 0.5),
    15.         (0, 0, 0.5, 0.5, 'reds', 1),
    16.         (0.5, 0, 1, 0.5, 'blues', 2)
    17.     )):
    19.             continue
    20.         data = state.rand(N, N) * scale
    21.         x, y = np.linspace(x0, x1, N + 1), np.linspace(y0, y1, N + 1)
    22.         m = ax.pcolormesh(
    23.             x, y, data, cmap=cmap,
    24.             levels=np.linspace(0, scale, 11)
    25.         )
    26.         ax.colorbar(m, loc='l', label=f'dataset #{i+1}')

    In ProPlot, colorbars and legends can be added to the edge of figures with the and proplot.figure.Figure.legend methods. Figure colorbars and legends are aligned between the edges of the subplot grid, rather than the figure bounds. As with , if you draw multiple colorbars or legends on the same side, they are stacked on top of each other.

    To draw a colorbar or legend alongside particular row(s) or column(s) of the subplot grid, use the row, rows, col, or cols keyword arguments. Pass an integer to draw the colorbar or legend beside a single row or column, or pass a tuple to draw it beside a range of rows or columns.

    1. import numpy as np
    2. fig, axs = plot.subplots(ncols=3, nrows=3, axwidth=1.4)
    3. state = np.random.RandomState(51423)
    4. m = axs.pcolormesh(
    5.     state.rand(20, 20), cmap='grays',
    6.     levels=np.linspace(0, 1, 11), extend='both'
    7. )[0]
    8. axs.format(
    9.     suptitle='Figure colorbars and legends demo', abc=True,
    10.     abcloc='l', abcstyle='a.', xlabel='xlabel', ylabel='ylabel'
    11. )
    12. fig.colorbar(m, label='column 1', ticks=0.5, loc='b', col=1)
    13. fig.colorbar(m, label='columns 2-3', ticks=0.2, loc='b', cols=(2, 3))
    14. fig.colorbar(m, label='stacked colorbar', ticks=0.1, loc='b', minorticks=0.05)
    15. fig.colorbar(m, label='colorbar with length <1', ticks=0.1, loc='r', length=0.7)
    1. [3]:
    1. <matplotlib.colorbar.Colorbar at 0x7fdd0aaeb880>

    _images/colorbars_legends_5_1.svg

    1. import proplot as plot
    2. import numpy as np
    3. fig, axs = plot.subplots(
    4.     ncols=2, nrows=2, axwidth=1.7,
    5.     share=0, wspace=0.3, order='F'
    6. )
    8. # Plot data
    9. data = (np.random.rand(50, 50) - 0.1).cumsum(axis=0)
    10. m = axs[:2].contourf(data, cmap='grays', extend='both')
    11. colors = plot.Colors('grays', 5)
    12. hs = []
    13. state = np.random.RandomState(51423)
    14. for abc, color in zip('ABCDEF', colors):
    15.     h = axs[2:].plot(state.rand(10), lw=3, color=color, label=f'line {abc}')
    16.     hs.extend(h[0])
    18. # Add colorbars and legends
    19. fig.colorbar(m[0], length=0.8, label='colorbar label', loc='b', col=1, locator=5)
    20. fig.colorbar(m[0], label='colorbar label', loc='l')
    21. fig.legend(hs, ncols=2, center=True, frame=False, loc='b', col=2)
    22. fig.legend(hs, ncols=1, label='legend label', frame=False, loc='r')
    23. axs.format(
    24.     suptitle='Figure colorbars and legends demo',
    25.     abc=True, abcloc='ul', abcstyle='A'
    26. )
    27. for ax, title in zip(
    28.     axs, ['2D dataset #1', '2D dataset #2', 'Line set #1', 'Line set #2']
    29. ):
    30.     ax.format(xlabel='xlabel', title=title)

    You can now draw colorbars from lists of colors or lists of artists by passing a list instead of a mappable object. Colorbar minor ticks are now much more robust, and the tick location and formatter arguments are passed through Locator and . The colorbar width and length can be changed with the width and length keyword args. Colorbar widths are now specified in physical units, which helps avoid colorbars that look “too skinny” or “too fat” and preserves the look of the figure when the figure size changes.

    1. [5]:
    2. import numpy as np
    3. fig, axs = plot.subplots(share=0, ncols=2, axwidth=2)
    5. # Colorbars from lines
    6. ax = axs[0]
    7. state = np.random.RandomState(51423)
    8. data = 1 + (state.rand(12, 10) - 0.45).cumsum(axis=0)
    9. cycle = plot.Cycle('algae')
    10.     data, lw=4, cycle=cycle, colorbar='lr',
    11.     colorbar_kw={'length': '8em', 'label': 'from lines'}
    12. )
    13. axs.colorbar(
    14.     hs, loc='t', values=np.arange(0, 10),
    15.     label='from lines', ticks=2
    16. )
    18. # Colorbars from a mappable
    19. ax = axs[1]
    20. m = ax.contourf(
    21.     data.T, extend='both', cmap='algae',
    22.     levels=plot.arange(0, 3, 0.5)
    23. )
    24. fig.colorbar(
    25.     m, length=1, loc='r', label='inside ticks',
    26.     tickloc='left'
    27. )
    28. ax.colorbar(
    29.     m, loc='ul', length=1, tickminor=True,
    30.     label='inset colorbar', alpha=0.5
    31. )
    32. axs.format(
    33.     suptitle='Colorbar formatting demo',
    34.     xlabel='xlabel', ylabel='ylabel', abovetop=False
    35. )

    _images/colorbars_legends_8_0.svg

    The proplot.figure.Figure.legend and methods are wrapped by legend_wrapper, which adds several new features.

    You can draw legends with centered legend rows, either by passing center=True or by passing list of lists of plot handles. This is accomplished by stacking multiple single-row, horizontally centered legends, then manually adding an encompassing legend frame. You can also modify legend text and handle properties with several keyword args, and switch between row-major and column-major order for legend entries with the order keyword arg (default is row-major).

    1. import proplot as plot
    2. import numpy as np
    3. plot.rc.cycle = '538'
    4. labels = ['a', 'bb', 'ccc', 'dddd', 'eeeee']
    5. fig, axs = plot.subplots(ncols=2, span=False, share=1, axwidth=2.3)
    6. hs1, hs2 = [], []
    8. # On-the-fly legends
    9. state = np.random.RandomState(51423)
    10. for i, label in enumerate(labels):
    11.     data = (state.rand(20) - 0.45).cumsum(axis=0)
    12.     h1 = axs[0].plot(
    13.         data, lw=4, label=label, legend='ul',
    14.         legend_kw={'order': 'F', 'title': 'column major'}
    15.     )
    16.     hs1.extend(h1)
    17.     h2 = axs[1].plot(
    18.         data, lw=4, label=label, legend='r', cycle='Set3',
    19.         legend_kw={'ncols': 1, 'frame': False, 'title': 'no frame'}
    20.     )
    21.     hs2.extend(h2)
    23. # Outer legends
    24. ax = axs[0]
    25. ax.legend(
    26.     hs1, loc='b', ncols=3, title='row major', order='C',
    27.     facecolor='gray2'
    28. )
    29. ax = axs[1]
    30. ax.legend(hs2, loc='b', ncols=3, center=True, title='centered rows')