如下所示:
import pandas as pd
path='F:/python/python数据分析与挖掘实战/图书配套数据、代码/chapter3/demo/data/catering_fish_congee.xls'
data=pd.read_excel(path,header=None,index_col=0)
data.index.name='日期'
data.columns=['销售额(元)']
xse=data['销售额(元)']
print(xse.max())
print(xse.min())
print(xse.max()-xse.min())
fanwei=list(range(0,4500,500))
fenzu=pd.cut(xse.values,fanwei,right=False)#分组区间,长度91
print(fenzu.codes)#标签
print(fenzu.categories)#分组区间,长度8
pinshu=fenzu.value_counts()#series,区间-个数
print(pinshu.index)
import matplotlib.pyplot as plt
pinshu.plot(kind='bar')
#plt.text(0,29,str(29))
qujian=pd.cut(xse,fanwei,right=False)
data['区间']=qujian.values
data.groupby('区间').median()
data.groupby('区间').mean()#每个区间平均数
pinshu_df=pd.DataFrame(pinshu,columns=['频数'])
pinshu_df['频率f']=pinshu_df / pinshu_df['频数'].sum()
pinshu_df['频率%']=pinshu_df['频率f'].map(lambda x:'%.2f%%'%(x*100))
pinshu_df['累计频率f']=pinshu_df['频率f'].cumsum()
pinshu_df['累计频率%']=pinshu_df['累计频率f'].map(lambda x:'%.4f%%'%(x*100))
In[158]: pinshu_df
Out[158]:
频数 频率f 频率% 累计频率f 累计频率%
[0, 500) 29 0.318681 31.87% 0.318681 31.8681%
[500, 1000) 20 0.219780 21.98% 0.538462 53.8462%
[1000, 1500) 12 0.131868 13.19% 0.670330 67.0330%
[1500, 2000) 12 0.131868 13.19% 0.802198 80.2198%
[2000, 2500) 8 0.087912 8.79% 0.890110 89.0110%
[2500, 3000) 3 0.032967 3.30% 0.923077 92.3077%
[3000, 3500) 4 0.043956 4.40% 0.967033 96.7033%
[3500, 4000) 3 0.032967 3.30% 1.000000 100.0000%
以上这篇pandas分区间,算频率的实例就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持脚本之家。