Python画柱状图

import numpy as np
def GetBarFrequency(arr, barTotal, minNum = None, maxNum = None):
    #没有自定义最大最小值则用数组计算最大最小值
    if not maxNum:
        maxNum = max(arr)
    if not minNum:
        minNum = min(arr)
    interval = (maxNum - minNum) / barTotal
    #定义多一个元素，后面就会发现它的作用
    barCntArr = np.zeros(barTotal + 1, dtype='int')
    for item in arr:
        barIndex = int((item - minNum) / interval)
        barCntArr[barIndex] += 1
    #我们肯定会有至少一个数刚好等于barTotal，所以数组会越界，上面数组定义了多一个元素，多出来的这个元素的统计值应该是归入第barTotal个元素即barCntArr[barTotal - 1]中
    barCntArr[barTotal - 1] += barCntArr[barTotal]
    #drop last one
    barCntArr = barCntArr[0: -1]
    coordinats = [minNum + i * interval for i in range(barTotal)]
    return (barCntArr, coordinats)

import numpy as np
import matplotlib.pyplot as plt

barTotal = 7
barCntArr1 = np.array([2, 3, 4, 1, 1, 2, 1])
barCntArr2 = np.array([1, 2, 3, 1, 1, 3, 2])

width = 0.35
ind = np.arange(barTotal)
fig, ax = plt.subplots()
rects1 = ax.bar(ind, barCntArr1, width, color='r')
rects2 = ax.bar(ind + width, barCntArr2, width, color='b')

ax.set_ylabel('y_label')
ax.set_title('title')
ax.set_xlabel('x_label')
ax.set_xticks(ind+width)
interval = 1.1
ax.set_xticklabels( [i * interval for i in range(7)] )
ax.legend( (rects1[0], rects2[0]), ('red', 'blue') )

plt.show()