Matplotlib: Stacked Bar Graphs

我在用matplotlib's stacked bar graph example 做bar统计图的时候，发现画出的图不能够准确匹配该项的值，几番查找原因，最终在Stack Overflow上找到了答案。以下给出一个 Stacked Bar Graphs demo。

import numpy as np

import matplotlib.pyplot as plt

N = 3

#('PledMis', 'PledFel', 'FoundFel')

Incarceration = np.array([115.3, 1.99,23.7] )

Probation = np.array([52.0, 45.9, 45.0])

Work = np.array([0, 0.3, 2.4])

Program= np.array([0, 0, 12])

ind = np.arange(N)    # the x locations for the groups

width = 0.5           # the width of the bars: can also be len(x) sequence

p1 = plt.bar(ind, Incarceration, width, color='r')

p2 = plt.bar(ind, Probation, width, color='y',bottom=Incarceration)

p3 = plt.bar(ind, Work, width, color='b',bottom= Probation+Incarceration)

p4 = plt.bar(ind, Program, width, color='g',bottom=Work+Probation+Incarceration)

plt.ylabel('Months')

plt.title('Time')

plt.xticks(ind+width/2, ('Found Guilty: Felony', 'Pled Guilty: Mis', 'Pled Guilty: Felony' ) )

plt.yticks(np.arange(10,200,10))

plt.legend ((p1[0], p2[0], p3[0], p4[0]), ('Incarceration', 'Probation','Work','Program' ))

plt.show()

图示结果：

如果我们做多个bar并且想要将之堆积起来，更有效的办法是用一个循环并且定义一个bottom自变量，代码如下：

colors = ('r', 'y', 'b', 'g')

data = (Incarceration, Probation, Work, Program)

bottom = np.zeros(N)

for elem, color in zip(data, colors):

    plt.bar(ind, elem, width, bottom=bottom, color=color)

    bottom += elem