How to reduce horizontal subplot spacing when tight_layout fails?

Question

I'm trying to create a single figure in matplotlib with several subplots of different row and column spans, using subplot2grid.

For example, the following code:

import matplotlib.pyplot as plt
import numpy as n

fig = plt.figure()
ax1 = plt.subplot2grid((2, 5), (0, 0), rowspan=2, colspan=2)
ax2 = plt.subplot2grid((2, 5), (0, 2))
ax2_2 = plt.subplot2grid((2, 5), (0, 3))
ax2_3 = plt.subplot2grid((2, 5), (0, 4))
ax3 = plt.subplot2grid((2, 5), (1, 2), colspan=3) #, sharex=ax2)

xs = n.linspace(0, 2*n.pi, 100)
ax1.plot(xs, n.sin(xs))
ax2.plot(xs, n.cos(xs))
ax2_2.plot(xs, n.tan(xs))
ax2_3.plot(xs, n.sin(xs))
ax3.plot(xs, n.cos(xs))


ax2.set_xticklabels([])
ax2.set_yticklabels([])
ax2_2.set_xticklabels([])
ax2_2.set_yticklabels([])
ax2_3.set_xticklabels([])
ax2_3.set_yticklabels([])

# This comes from a separate attempt to make the ticks invisible, but seems to make no change.
plt.setp(ax2.get_xticklabels(), visible=False)
plt.setp(ax2.get_yticklabels(), visible=False)
plt.setp(ax2_2.get_xticklabels(), visible=False)
plt.setp(ax2_2.get_yticklabels(), visible=False)
plt.setp(ax2_3.get_xticklabels(), visible=False)
plt.setp(ax2_3.get_yticklabels(), visible=False)


fig.set_size_inches(6.5, 2.85)
fig.tight_layout()
# I tried several things here - subplots_adjust improves the vertical spacing a little, but not the horizontal.
fig.subplots_adjust(hspace=0.1)

fig.savefig('test.png', dpi=100)

This makes a figure like:

The problem is the 3 single-cell subplots in the top right, each of which has only about 70% of the width that would fit. tight_layout otherwise takes care of the spacing nicely, but I can't work out why this space isn't filled, or how to work around it.

So...does anyone know how to do this, making the subplots take the full available width?

Answer 1

当使用subplots_adjust() ， hspace参数将调整垂直空间，而wspace参数将调整水平空间，因此您可能需要的是：

fig.subplots_adjust(wspace=0.1)