每个子图规格的标题（子图内的子图）

Posted 2021-04-12

我想制作一个具有2x2外部布局并在这2x3布局的每个布局内的图。我设法使用gridSpec使一切正常，但是似乎无法为每个subplotSpec分配一个标题（有2x2 subplotSpec）。

Row titles for matplotlib subplot原则上，这应该给我解决方案，但是我想知道是否存在使用gridSpec而不隐藏白框的方法。

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec

netdict = ["chain","V"]
sigdict = ["no","with"]

fig = plt.figure(figsize=(10, 8))
outer = gridspec.GridSpec(2, 2, wspace=0.3, hspace=0.3) # make a 2x2 outer frame

for ni, networktype in enumerate(netdict):
    for si, sigtype in enumerate(sigdict):
        fig.add_subplot(outer[ni*2+si])
        plt.title(networktype+ " " + sigtype) # title for each outer frame
        plt._frameon=False # this should make the unnecessary white frame go away
        inner = gridspec.GridSpecFromSubplotSpec(2, 3, # make 2x3 inner frame
                        subplot_spec=outer[ni*2+si], wspace=0.4, hspace=0.4)
        for ti,tau in enumerate([5.5,12.5]):
            for u, xy_delay in enumerate([0,tau,2*tau]):
                ax = plt.Subplot(fig, inner[ti*3+u])
                ax.plot(np.sin(np.linspace(0,10,0.5))
                ax.set(title=r"$\tau_u={}$".format(xy_delay))
                fig.add_subplot(ax)
plt.show()

答案

只是偶然发现了这一点，因为我不得不做类似的事情。我发现一个简单的解决方案是为外部网格创建另一个子图，然后简单地将其轴关闭：

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec

fig = plt.figure(figsize=(12, 6))
outer = gridspec.GridSpec(1, 2, wspace=0.2, hspace=0.2)

for i, layer in enumerate(layer_list):
    # create inner plots
    inner = gridspec.GridSpecFromSubplotSpec(2, 2,
                    subplot_spec=outer[i], wspace=0.1, hspace=0.1)
    # set outer titles
    ax = plt.Subplot(fig, outer[i])
    ax.set_title("Layer {}".format(layer))
    ax.axis('off')
    fig.add_subplot(ax)
    for j, filter_pos in enumerate(filter_pos_list):
        # create image plot
        ax = plt.Subplot(fig, inner[j])
        # load image
        img = ...
        ax.imshow(img)
        # set inner title
        ax.set_title("Filter {}".format(filter_pos))
        ax.set_xticks([])
        ax.set_yticks([])
        fig.add_subplot(ax)
fig.show()