Question

如何为多索引数据框的行建立索引

import pandas as pd
import numpy as np
np.random.seed(0)
tuples = list(zip(*[['bar', 'bar', 'baz', 'baz'],['one', 'two', 'one', 'two']]))
idx = pd.MultiIndex.from_tuples(tuples, names=['first', 'second'])
df = pd.DataFrame(np.random.randn(4, 2), index=idx, columns=['A', 'B'])
print(df)
                     A         B
first second
bar   one     1.764052  0.400157
      two     0.978738  2.240893
baz   one     1.867558 -0.977278
      two     0.950088 -0.151357

使用第二个数据框的列

idxDf = pd.DataFrame({'first':['bar','baz'],'second':['one','two']})
print(idxDf)
  first second
0   bar    one
1   baz    two

使得结果数据帧为

first second
bar   one     1.764052  0.400157
baz   two     0.950088 -0.151357

？

很显然，df[idxDf['first','second']]不起作用。

Answer 1

将DataFrame.merge与DataFrame.reset_index和DataFrame.set_index结合使用：

print (df.reset_index().merge(idxDf, on=['first','second']).set_index(['first','second']))
                     A         B
first second                    
bar   one     1.764052  0.400157
baz   two     0.950088 -0.151357

或DataFrame.set_index：

print (df.merge(idxDf, 
                left_index=True, 
                right_on=['first','second']).set_index(['first','second']))
                     A         B
first second                    
bar   one     1.764052  0.400157
baz   two     0.950088 -0.151357

或在merge之前的DataFrame.set_index：

print (df.merge(idxDf.set_index(['first','second']), 
                left_index=True, 
                right_index=True))
                     A         B
first second                    
bar   one     1.764052  0.400157
baz   two     0.950088 -0.151357

使用数据框的列值为多索引数据框的行编制索引

1 个答案: