In [1]:

    

import numpy as np 
import pandas as pd
from pandas import Series,DataFrame


    

In [2]:

    

dframe1 = DataFrame({'key': ['X', 'Z', 'Y', 'Z', 'X', 'X'], 'data_set_1': np.arange(6)})

dframe1


    

    
Out[2]:






  

    

      

      
data_set_1
      
key
    
  
  

    

      
0
      
0
      
X
    
    

      
1
      
1
      
Z
    
    

      
2
      
2
      
Y
    
    

      
3
      
3
      
Z
    
    

      
4
      
4
      
X
    
    

      
5
      
5
      
X
    
  

In [5]:

    

dframe2 = DataFrame({'key': ['Q', 'Y', 'Z'], 'data_set_2': np.arange(3) + 1})

dframe2


    

    
Out[5]:






  

    

      

      
data_set_2
      
key
    
  
  

    

      
0
      
1
      
Q
    
    

      
1
      
2
      
Y
    
    

      
2
      
3
      
Z
    
  

In [6]:

    

pd.merge(dframe1, dframe2)


    

    
Out[6]:






  

    

      

      
data_set_1
      
key
      
data_set_2
    
  
  

    

      
0
      
1
      
Z
      
3
    
    

      
1
      
3
      
Z
      
3
    
    

      
2
      
2
      
Y
      
2
    
  

In [7]:

    

pd.merge(dframe1, dframe2, on='key')


    

    
Out[7]:






  

    

      

      
data_set_1
      
key
      
data_set_2
    
  
  

    

      
0
      
1
      
Z
      
3
    
    

      
1
      
3
      
Z
      
3
    
    

      
2
      
2
      
Y
      
2
    
  

In [8]:

    

pd.merge(dframe1, dframe2, on='key', how='left')


    

    
Out[8]:






  

    

      

      
data_set_1
      
key
      
data_set_2
    
  
  

    

      
0
      
0
      
X
      
NaN
    
    

      
1
      
1
      
Z
      
3.0
    
    

      
2
      
2
      
Y
      
2.0
    
    

      
3
      
3
      
Z
      
3.0
    
    

      
4
      
4
      
X
      
NaN
    
    

      
5
      
5
      
X
      
NaN
    
  

In [9]:

    

pd.merge(dframe1, dframe2, on='key', how='right')


    

    
Out[9]:






  

    

      

      
data_set_1
      
key
      
data_set_2
    
  
  

    

      
0
      
1.0
      
Z
      
3
    
    

      
1
      
3.0
      
Z
      
3
    
    

      
2
      
2.0
      
Y
      
2
    
    

      
3
      
NaN
      
Q
      
1
    
  

In [10]:

    

pd.merge(dframe1, dframe2, on='key', how='outer')


    

    
Out[10]:






  

    

      

      
data_set_1
      
key
      
data_set_2
    
  
  

    

      
0
      
0.0
      
X
      
NaN
    
    

      
1
      
4.0
      
X
      
NaN
    
    

      
2
      
5.0
      
X
      
NaN
    
    

      
3
      
1.0
      
Z
      
3.0
    
    

      
4
      
3.0
      
Z
      
3.0
    
    

      
5
      
2.0
      
Y
      
2.0
    
    

      
6
      
NaN
      
Q
      
1.0
    
  

In [11]:

    

dframe3 = DataFrame({'key': ['X', 'X', 'X', 'Y', 'Z', 'Z'], 'data_set_3': range(6)})


    

In [12]:

    

dframe4 = DataFrame({'key': ['Y', 'Y', 'X', 'X', 'Z'], 'data_set_4': range(5)})


    

In [13]:

    

pd.merge(dframe3, dframe4)


    

    
Out[13]:






  

    

      

      
data_set_3
      
key
      
data_set_4
    
  
  

    

      
0
      
0
      
X
      
2
    
    

      
1
      
0
      
X
      
3
    
    

      
2
      
1
      
X
      
2
    
    

      
3
      
1
      
X
      
3
    
    

      
4
      
2
      
X
      
2
    
    

      
5
      
2
      
X
      
3
    
    

      
6
      
3
      
Y
      
0
    
    

      
7
      
3
      
Y
      
1
    
    

      
8
      
4
      
Z
      
4
    
    

      
9
      
5
      
Z
      
4
    
  

In [14]:

    

df_left = DataFrame({'key1': ['SF', 'SF', 'LA'], 
                     'key2': ['one', 'two', 'three'],
                     'left_data': [10,20,30]})


    

In [15]:

    

df_right = DataFrame({'key1': ['SF', 'SF', 'LA', 'LA'], 
                     'key2': ['one', 'one', 'one', 'three'],
                     'left_data': [40,50,60,79]})


    

In [16]:

    

pd.merge(df_left, df_right, on=['key1', 'key2'], how='outer')


    

    
Out[16]:






  

    

      

      
key1
      
key2
      
left_data_x
      
left_data_y
    
  
  

    

      
0
      
SF
      
one
      
10.0
      
40.0
    
    

      
1
      
SF
      
one
      
10.0
      
50.0
    
    

      
2
      
SF
      
two
      
20.0
      
NaN
    
    

      
3
      
LA
      
three
      
30.0
      
79.0
    
    

      
4
      
LA
      
one
      
NaN
      
60.0
    
  

In [17]:

    

pd.merge(df_left, df_right, on=['key1'], how='outer')


    

    
Out[17]:






  

    

      

      
key1
      
key2_x
      
left_data_x
      
key2_y
      
left_data_y
    
  
  

    

      
0
      
SF
      
one
      
10
      
one
      
40
    
    

      
1
      
SF
      
one
      
10
      
one
      
50
    
    

      
2
      
SF
      
two
      
20
      
one
      
40
    
    

      
3
      
SF
      
two
      
20
      
one
      
50
    
    

      
4
      
LA
      
three
      
30
      
one
      
60
    
    

      
5
      
LA
      
three
      
30
      
three
      
79
    
  

In [18]:

    

pd.merge(df_left, df_right, on=['key1'], how='outer', suffixes=('_lefty', '_righty'))


    

    
Out[18]:






  

    

      

      
key1
      
key2_lefty
      
left_data_lefty
      
key2_righty
      
left_data_righty
    
  
  

    

      
0
      
SF
      
one
      
10
      
one
      
40
    
    

      
1
      
SF
      
one
      
10
      
one
      
50
    
    

      
2
      
SF
      
two
      
20
      
one
      
40
    
    

      
3
      
SF
      
two
      
20
      
one
      
50
    
    

      
4
      
LA
      
three
      
30
      
one
      
60
    
    

      
5
      
LA
      
three
      
30
      
three
      
79
    
  

In [ ]: