import pandas as pd
path='F:/python/python數(shù)據(jù)分析與挖掘?qū)崙?zhàn)/圖書配套數(shù)據(jù)、代碼/chapter3/demo/data/catering_fish_congee.xls'
data=pd.read_excel(path,header=None,index_col=0)
data.index.name='日期'
data.columns=['銷售額(元)']
 
xse=data['銷售額(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)#分組區(qū)間,長(zhǎng)度91
print(fenzu.codes)#標(biāo)簽
print(fenzu.categories)#分組區(qū)間，長(zhǎng)度8
pinshu=fenzu.value_counts()#series,區(qū)間-個(gè)數(shù)
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['區(qū)間']=qujian.values
data.groupby('區(qū)間').median()
data.groupby('區(qū)間').mean()#每個(gè)區(qū)間平均數(shù)
 
pinshu_df=pd.DataFrame(pinshu,columns=['頻數(shù)'])
pinshu_df['頻率f']=pinshu_df / pinshu_df['頻數(shù)'].sum()
pinshu_df['頻率%']=pinshu_df['頻率f'].map(lambda x:'%.2f%%'%(x*100))
 
pinshu_df['累計(jì)頻率f']=pinshu_df['頻率f'].cumsum()
pinshu_df['累計(jì)頻率%']=pinshu_df['累計(jì)頻率f'].map(lambda x:'%.4f%%'%(x*100))
 
In[158]: pinshu_df
Out[158]: 
       頻數(shù)    頻率f   頻率%   累計(jì)頻率f   累計(jì)頻率%
[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%