使用 pd.merge 来实现
on 表示查询的 columns，如果都有 id，那么这是很好的区别项，找到 id 相同的进行merge。
>>> import numpy as np
>>> import pandas as pd
>>> data1 = {
'one': pd.Series([1,2,3]),
'two': pd.Series([11,22,33])
}
>>> df1 = pd.DataFrame(data = data1)
>>> df1
one two
0 1 11
1 2 22
2 3 33
>>> data2 = {
'one': pd.Series([1,2,3,4,5,6]),
'two': pd.Series([11,22,33]),
'three': pd.Series([111,222,333]),
'four': pd.Series([1111,2222,3333,4444,5555,6666])
}
>>> df2 = pd.DataFrame(data = data2)
>>> df2
one two three four
0 1 11.0 111.0 1111
1 2 22.0 222.0 2222
2 3 33.0 333.0 3333
3 4 NaN NaN 4444
4 5 NaN NaN 5555
5 6 NaN NaN 6666
>>> df2[df2['one']<3]
one two three four
0 1 11.0 111.0 1111
1 2 22.0 222.0 2222
>>> df = pd.merge(df1, df2, how='inner')
>>> df
one two three four
0 1 11 111.0 1111
1 2 22 222.0 2222
2 3 33 333.0 3333
>>> df1
one two
0 1 11
1 2 22
2 3 33
>>> df2
one two three four
0 1 11.0 111.0 1111
1 2 22.0 222.0 2222
2 3 33.0 333.0 3333
3 4 NaN NaN 4444
4 5 NaN NaN 5555
5 6 NaN NaN 6666
>>> pd.merge(df1, df2, how='inner')
one two three four
0 1 11 111.0 1111
1 2 22 222.0 2222
2 3 33 333.0 3333
>>> pd.merge(df2, df1, how='inner')
one two three four
0 1 11.0 111.0 1111
1 2 22.0 222.0 2222
2 3 33.0 333.0 3333
>>> five = pd.Series([1,2,3,4,5,6])
>>> df2['five'] = five
>>> df2
one two three four five
0 1 11.0 111.0 1111 1
1 2 22.0 222.0 2222 2
2 3 33.0 333.0 3333 3
3 4 NaN NaN 4444 4
4 5 NaN NaN 5555 5
5 6 NaN NaN 6666 6
>>> df1
one two
0 1 11
1 2 22
2 3 33
>>> pd.merge(df2, df1, how='inner')
one two three four five
0 1 11.0 111.0 1111 1
1 2 22.0 222.0 2222 2
2 3 33.0 333.0 3333 3
>>> pd.merge(df1, df2, how='inner')
one two three four five
0 1 11 111.0 1111 1
1 2 22 222.0 2222 2
2 3 33 333.0 3333 3
>>> df1
one two
0 1 11
1 2 22
2 3 33
>>> df2
one two three four five
0 1 11.0 111.0 1111 1
1 2 22.0 222.0 2222 2
2 3 33.0 333.0 3333 3
3 4 NaN NaN 4444 4
4 5 NaN NaN 5555 5
5 6 NaN NaN 6666 6
>>> six = pd.Series([-1, -2, -3])
>>> df1['six'] = six
>>> df1
one two six
0 1 11 -1
1 2 22 -2
2 3 33 -3
>>> df2
one two three four five
0 1 11.0 111.0 1111 1
1 2 22.0 222.0 2222 2
2 3 33.0 333.0 3333 3
3 4 NaN NaN 4444 4
4 5 NaN NaN 5555 5
5 6 NaN NaN 6666 6
>>> pd.merge(df1, df2, how='inner')
one two six three four five
0 1 11 -1 111.0 1111 1
1 2 22 -2 222.0 2222 2
2 3 33 -3 333.0 3333 3
>>> pd.merge(df2, df1, how='inner')
one two three four five six
0 1 11.0 111.0 1111 1 -1
1 2 22.0 222.0 2222 2 -2
2 3 33.0 333.0 3333 3 -3