pandas 设置多重索引_pandas基础(2)_多重索引

1：多重索引的构造

>>> #下面显示构造pd.MultiIndex

>>> df1=DataFrame(np.random.randint(0,150,size=(6,3)),columns=['java','html5','python'])

>>> import pandas as pd

>>> df1=DataFrame(np.random.randint(0,150,size=(6,3)),columns=['java','html5','python'],index=pd.MultiIndex.from_arrays([['张三','张三','侯少','侯少','a','a'],['M','E','M','E','M','E']]))

>>> df1#因为Python自身的原因，对汉字的识别不是太好，所以汉字被？代替了

java  html5  python

???? M     2     13      76

E   141     67      84

M   116     83       8

E    70    118     125

a    M    74      0      76

E   111     31       8

>>> #使用元组tuple创建

df2=DataFrame(np.random.randint(0,150,size=(6,3)),columns=['java','html','python'],index=pd.MultiIndex.from_tuples([('a','1'),('a','11'),('b','1'),('b','11'),('c','1'),('c','11')]))

>>> df2

java  html  python

a 1     32   144      99

11   104   101      16

b 1     93    98      41

11    59    30      45

c 1     91    17     149

11     9    28      59

>>> #使用product

df2=DataFrame(np.random.randint(0,150,size=(6,3)),columns=['java','html','python'],index=pd.MultiIndex.from_product([['zhangsan ','lisi','wangwu'],['mid','end']]))

>>> df2

java  html  python

zhangsan  mid    50   128      54

end     3     4      91

lisi      mid     4    93     110

end   116   123     122

wangwu    mid    88    25      54

end    48   146      57

>>> #对dataFrame同样可以设置成多重索引

df2=DataFrame(np.random.randint(0,150,size=(3,6)),columns=pd.MultiIndex.from_product([['java','html','python'],['mid','end']]),index=['张三','李四','王五'])

>>> df2

java     html     python

mid end  mid end    mid  end

????   33  38  112  70    113  110

????   29  46  132  91    117  128

????   73  56  118  82    132   39

>>>

>>> df2['java','mid']#查询某一列

????    33

????    29

????    73

Name: (java, mid), dtype: int32

>>> s['zhangsan':'lisi']#其实就是一个Series

Series([], dtype: int64)

>>> s.iloc[0:3]

a  0    1

1    2

b  0    3

dtype: int64

>>> #切片

>>> df2['张三':'王五']

java     html     python

mid end  mid end    mid  end

????   33  38  112  70    113  110

????   29  46  132  91    117  128

????   73  56  118  82    132   39

>>>df2.iloc[0:4]#推荐使用

Df2[‘张三’，‘期中’]和df2.loc[‘张三’].loc[‘期中’]

#如何一级索引有多个，对二级索引会遇到问题，也就是说，无法直接对二级进行索引

必须把二级索引变成一级索引才可以进行索引

>>> df2.stack()

html  java  python

???? end    70    38     110

mid   112    33     113

end    91    46     128

mid   132    29     117

end    82    56      39

mid   118    73     132

>>> #stack =堆----》行

end  mid

???? html     70  112

java     38   33

python  110  113

html     91  132

java     46   29

python  128  117

html     82  118

java     56   73

python   39  132

>>> #默认为-1

2：多重索引的计算

>>> df2

java     html     python

mid end  mid end    mid  end

????   33  38  112  70    113  110

????   29  46  132  91    117  128

????   73  56  118  82    132   39

>>> df1.sum()

java      514

html5     312

python    377

dtype: int64

>>> df1.sum(axis=0)

java      514

html5     312

python    377

dtype: int64

>>> df1.sum(axis=1)#对列

????  M     91

E    292

M    207

E    313

a     M    150

E    150

dtype: int64

>>> df1.sum(axis=1)#对列求和，得到每行的和

????  M     91

E    292

M    207

E    313

a     M    150

E    150

dtype: int64

>>> df1.std

???? M     2     13      76

E   141     67      84

M   116     83       8

E    70    118     125

a    M    74      0      76

E   111     31       8>

>>> #求方差

>>> df1.std(axis=1)

????  M    39.929104

E    38.759945

M    55.344376

E    29.938827

a     M    43.312816

E    54.064776

dtype: float64

>>> df1.max()

java      141

html5     118

python    125

dtype: int32

3多重索引的拼接

>>> nd = np.random.randint(0,10,size=(3,3))

>>> nd

array([[9, 9, 4],

[7, 2, 4],

[1, 6, 1]])

>>> np.concatenate ((nd,nd),axis=0)#在列方向就行拼接

array([[9, 9, 4],

[7, 2, 4],

[1, 6, 1],

[9, 9, 4],

[7, 2, 4],

[1, 6, 1]])

>>> np.concatenate ([nd,nd],axis=1)#在行方向进行拼接

array([[9, 9, 4, 9, 9, 4],

[7, 2, 4, 7, 2, 4],

[1, 6, 1, 1, 6, 1]])

>>> def make_df(cols,inds):

data = {c:[c+str(i) for i in cols]for c in cols}

return DataFrame(data,index=inds,columns=cols)

>>> make_df(['A','B'],[1,2])

A   B

1  AA  BA

2  AB  BB

>>> df1=make_df(list('AB'),[0,1])

>>> df2=make_df(list('AB'),[2,3])

>>> pd.concat ([df1,df2])#默认在列方向进行拼接

A   B

0  AA  BA

1  AB  BB

2  AA  BA

3  AB  BB

>>> #优先增加行数

>>> pd.concat ((df1,df2),axis=1)

A    B    A    B

0   AA   BA  NaN  NaN

1   AB   BB  NaN  NaN

2  NaN  NaN   AA   BA

3  NaN  NaN   AB   BB

>>> #注意index在级联时可以重复

3)

>>> #列名可以相同但是不建议

>>> df3= make_df(list('AB'),[0,1])

>>> df4=make_df(list('VB'),[1,2])

>>> pd.concat((df3,df4))#只能传入一个参数

A   B    V

0   AA  BA  NaN

1   AB  BB  NaN

1  NaN  BV   VV

2  NaN  BB   VB

>>> #3种连接方式

>>> #1：外连接：补NaN(默认模式)

>>> df1= make_df(list('AB'),[1,3])

>>> df2= make_df(list('AB'),[2,4])

>>> df2= make_df(list('BC'),[2,4])

>>> pd.concat ([df1,df2],join='inner')#连接都有的部分

B

1  BA

3  BB

2  BB

4  BC

>>> pd.concat ([df1,df2],join='outer')

A   B    C

1   AA  BA  NaN

3   AB  BB  NaN

2  NaN  BB   CB

4  NaN  BC   CC

>>> #内连接只连接匹配项

>>> #3：连接指定轴join_axes所以CDF的F便不显示了

>>> df3= make_df(list('ACD'),[0,1,2])

>>> df4= make_df(list('CDF'),[3,4,5])

>>> pd.concat([df3,df4],join_axes=[df3.columns])

A   C   D

0   AA  CA  DA

1   AC  CC  DC

2   AD  CD  DD

3  NaN  CC  DC

4  NaN  CD  DD

5  NaN  CF  DF

>>> #join_axes 某一个DataFrame列索引为新的列索引值

>>> #3使用append()函数添加

>>> #concat方法属于pandas

>>> #append()在后面添加

>>> #concat([df1,df2])

>>> #df1.append(df2)

>>> #merge与concat的区别是，merge需要依据某一共同的行或列来进行合并

>>> #使用pd.merge()合并时，会自动根据两者相同column名称的那一属性，作为key来进行合并，注意每一列的顺序不要求一致

>>> #一对一合并

>>> df1 = DataFrame({'employee':['po','sara','danis'],'group':['sail','counting','marcketing']})

>>> df2 = DataFrame({'employee':['po','sara','danis'],'work_time':[2,3,1]})

>>> df1

employee       group

0       po        sail

1     sara    counting

2    danis  marcketing

>>> df2

employee  work_time

0       po          2

1     sara          3

2    danis          1

>>> pd.merge (df1,df2)

employee       group  work_time

0       po        sail          2

1     sara    counting          3

2    danis  marcketing          1

>>> pd.concat([df1,df2])

employee       group  work_time

0       po        sail        NaN

1     sara    counting        NaN

2    danis  marcketing        NaN

0       po         NaN        2.0

1     sara         NaN        3.0

2    danis         NaN        1.0

>>> df3 = DataFrame({'employee':['po','sara','liulei'],'work_time':[2,3,1]})

>>> pd.merge(df1,df3)

employee     group  work_time

0       po      sail          2

1     sara  counting          3

>>> #merge只合并相同属性里面都有的项

>>> #下面是merge的多对一的合并

>>> df1 = DataFrame({'employee':['po','sara','danis'],'work_time':[2,3,1]})

>>> df2 = DataFrame({'employee':['po','po','danis'],'group':['sail','counting','marcketing']})

>>> pd.merge(df1,df2)

employee  work_time       group

0       po          2        sail

1       po          2    counting

2    danis          1  marcketing

>>> #出现了两个po

>>> #下面是多对多的合并

>>> df1 = DataFrame({'employee':['po','sara','danis'],'group':['sail','counting','marcketing']})

>>> df1 = DataFrame({'employee':['po','po','danis'],'group':['sail','counting','marcketing']})

>>> df2 = DataFrame({'employee':['po','po','danis'],'work_time':[2,3,1]})

>>> pd.merge(df1,df2)

employee       group  work_time

0       po        sail          2

1       po        sail          3

2       po    counting          2

3       po    counting          3

4    danis  marcketing          1

>>> #1*2*2的模式

>>> #使用merge多对多可以来处理重名等数据的情况

>>> df3= DataFrame({'employee':['po','Summer','Flower'],'group':['sail','marking','serch'],'WorkTime':[1,2,3]})

>>> df4= DataFrame({'employee':['po','Summer','Flower'],'group':['sail','marking','serch'],'salary':[12000,20000,10002]})

>>> df3

WorkTime employee    group

0         1       po     sail

1         2   Summer  marking

2         3   Flower    serch

>>> df4

employee    group  salary

0       po     sail   12000

1   Summer  marking   20000

2   Flower    serch   10002

>>> pd.merge(df3,df4)

WorkTime employee    group  salary

0         1       po     sail   12000

1         2   Summer  marking   20000

2         3   Flower    serch   10002

>>> df3= DataFrame({'employee':['po','Winter','Flower'],'group':['marketing','marking','serch'],'WorkTime':[1,2,3]})

>>> pd.merge(df3,df4)

WorkTime employee  group  salary

0         3   Flower  serch   10002

>>> pd.merge(df3,df4,on='employee')

WorkTime employee    group_x group_y  salary

0         1       po  marketing    sail   12000

1         3   Flower      serch   serch   10002

>>> #出现两行数据的原因是指定了employee相同就可以合并

>> pd.merge(df3,df4,on='group')

WorkTime employee_x    group employee_y  salary

0         2     Winter  marking     Summer   20000

1         3     Flower    serch     Flower   10002

>>> pd.merge(df3,df4,on='group',suffixes=['_A','_B'])

WorkTime employee_A    group employee_B  salary

0         2     Winter  marking     Summer   20000

1         3     Flower    serch     Flower   10002

>>> df3= DataFrame({'employee':['po','Winter','Flower'],'group':['marketing','marking','serch'],'WorkTime':[1,2,3]})

>>> df4= DataFrame({'employer':['po','Summer','Flower'],'group':['sail','marking','serch'],'salary':[12000,20000,10002]})

>>> pd.merge(df3,df4)

WorkTime employee    group employer  salary

0         2   Winter  marking   Summer   20000

1         3   Flower    serch   Flower   10002

>>> pd.merge(df3,df4,left_on='employee',right_on='employer')

WorkTime employee    group_x employer group_y  salary

0         1       po  marketing       po    sail   12000

1         3   Flower      serch   Flower   serch   10002

>>> #df3主键key为employee和df4主键为employer,两者不同但又想相互匹配时，可以指定前者的left_on为employee和后者的right_on为employer这时两者可以进行匹配

>>> #内合并与外合并

>>> #内合并只保留两者都有的数据

>>> df1=DataFrame({'age':[18,22,33],'height':[175,169,180]})

>>> df1=DataFrame({'age':[18,23,32],'height':[175,169,180]})

>>> df2=DataFrame({'age':[18,22,33],'weight':[175,169,180]})

>>> pd.merge(df1,df2)

age  height  weight

0   18     175     175

>>> pd.merge(df1,df2,how='outer')

age  height  weight

0   18   175.0   175.0

1   23   169.0     NaN

2   32   180.0     NaN

3   22     NaN   169.0

4   33     NaN   180.0

>>> #默认为内合并，通过how可以指定合并类型

>>>

>>> pd.merge(df1,df2,how='left')

age  height  weight

0   18     175   175.0

1   23     169     NaN

2   32     180     NaN

>>> pd.merge(df1,df2,how='right')

age  height  weight

0   18   175.0     175

1   22     NaN     169

2   33     NaN     180

>>> #left保留前者的数据，right保留后者数据

>>> #left保留前者df1的数据，right保留后者df2数据

>>> #下面是列冲突

>>> df3= DataFrame({'employee':['po','Winter','Flower'],'group':['marketing','marking','serch'],'WorkTime':[1,2,3]})

>>> df4= DataFrame({'employee':['po','Summer','Flower'],'group':['sail','marking','serch'],'salary':[12000,20000,10002]})

>>> pd.merge(df3,df4)

WorkTime employee  group  salary

0         3   Flower  serch   10002

>>> pd.merge(df3,df4,on='employee',suffixes=['_李','_王'])

WorkTime employee   group_?? group_??  salary

0         1       po  marketing     sail   12000

1         3   Flower      serch    serch   10002

>>> #因为两者的employee和group相同，当指定employee为主键时，suffixes修改的就是group

4：总结：

多重索引也是pandas里非常重要的知识点，要牢牢掌握