Matplotlib：在条形图中调整错误栏的z级别

Question

import pylab as pl
import numpy as np

x = np.arange(4)
y1, y2 = [1,2,1,1], [2,3,1,1.5]


pl.bar(x+0.2,y2, width=0.45, color='g')
pl.errorbar(x+0.4,y2,fmt=None, yerr=0.75, ecolor='r', 
            lw=2, capsize=10., mew = 3)

pl.bar(x,y1,width=0.45)
pl.errorbar(x+0.2,y1,fmt=None, yerr=0.5, ecolor='r', 
            lw=2, capsize=10., mew = 3)

pl.savefig('err.png')

产生

我希望绿色值的错误栏由蓝色条 覆盖 。

我认为调整绘图的z水平应该达到（这就是为什么我首先单独使用.bar和.errorbar）：

pl.bar(x+0.2,y2, width=0.45, color='g', zorder=1)
pl.errorbar(x+0.4,y2,fmt=None, yerr=0.75, ecolor='r', 
            lw=2, capsize=10., mew = 3, zorder=1)

pl.bar(x,y1,width=0.45, zorder=2)
pl.errorbar(x+0.2,y1,fmt=None, yerr=0.5, ecolor='r', 
            lw=2, capsize=10., mew = 3, zorder=2)

这会给

我无法找到有效的zorders组合。如何在Matplotlib条形图中正确调整错误栏的z级别？

Answer 1

我认为你必须分别设置上限的z顺序（它们是pylab.errorbar返回的三个对象之一：

import pylab as pl
import numpy as np

x = np.arange(4)
y1, y2 = [1,2,1,1], [2,3,1,1.5]


pl.bar(x+0.2,y2, width=0.45, color='g', zorder=1)
_, caplines, _ = pl.errorbar(x+0.4,y2,fmt=None, yerr=0.75, ecolor='r', 
            lw=2, capsize=10., mew = 3, zorder=2)

pl.bar(x,y1,width=0.45, zorder=3)
pl.errorbar(x+0.2,y1,fmt=None, yerr=0.5, ecolor='r', 
            lw=2, capsize=10., mew = 3, zorder=4)

for capline in caplines:
    capline.set_zorder(2)

pl.savefig('err.png')