pythonmatplotlib

How to fill spaces between subplots with a color in Matplotlib?

With the following code :

nb_vars=4

fig, axs = plt.subplots(4,4,figsize=(8,8), gridspec_kw = {'wspace':0.20, 'hspace':0.20}, dpi= 100)
for i_ax in axs:
    for ii_ax in i_ax:
        ii_ax.set_yticklabels([])
for i_ax in axs:
    for ii_ax in i_ax:
        ii_ax.set_xticklabels([])

The space between the subplots is white. How is it possible to colour them ? And with different colors ? See for example this figure : enter image description here

Solution

  • You could add patches in between the axes:

    from matplotlib import patches

nb_vars=4

# colors between two axes in a row
r_colors = [['#CC0000', '#CC0000', '#CC0000'],
            ['#0293D8', '#0293D8', '#0293D8'],
            ['#FF8E00', '#FF8E00', '#FF8E00'],
            ['#ABB402', '#ABB402', '#ABB402'],
           ]

# colors between two axes in a column
c_colors = [['#CC0000', '#0293D8', '#FF8E00', '#ABB402'],
            ['#CC0000', '#0293D8', '#FF8E00', '#ABB402'],
            ['#CC0000', '#0293D8', '#FF8E00', '#ABB402'],
           ]

fig, axs = plt.subplots(4, 4, figsize=(4, 4),
                        gridspec_kw = {'wspace':0.20, 'hspace':0.20}, dpi= 100)
h, w = axs.shape

for r, i_ax in enumerate(axs):
    for c, ii_ax in enumerate(i_ax):
        ii_ax.set_yticklabels([])
        ii_ax.set_xticklabels([])
        ii_ax.plot([-r, r], [-c, c]) # plot dummy line for demo
        bbox = ii_ax.get_position()
        if r+1 < h:
            ii_ax.add_patch(patches.Rectangle((bbox.x0, bbox.y0),
                                              bbox.width, -0.2,
                                              facecolor=c_colors[r][c],
                                              zorder=-1, clip_on=False,
                                              transform=fig.transFigure, figure=fig
                                             ))
        if c+1 < w:
            ii_ax.add_patch(patches.Rectangle((bbox.x1, bbox.y0),
                                              0.2, bbox.height,
                                              facecolor=r_colors[r][c],
                                              zorder=-1, clip_on=False,
                                              transform=fig.transFigure, figure=fig
                                             ))

    Output:

    enter image description here