八、 改变形状

详情请参阅 层次索引 和 改变形状。

Stack

  1. In [95]: tuples = list(zip(*[['bar', 'bar', 'baz', 'baz',
  2.    ....:                      'foo', 'foo', 'qux', 'qux'],
  3.    ....:                     ['one', 'two', 'one', 'two',
  4.    ....:                      'one', 'two', 'one', 'two']]))
  5.    ....: 
  7. In [96]: index = pd.MultiIndex.from_tuples(tuples, names=['first', 'second'])
  9. In [97]: df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=['A', 'B'])
  11. In [98]: df2 = df[:4]
  13. In [99]: df2
  14. Out[99]: 
  15.                      A         B
  16. first second                    
  17. bar   one     0.029399 -0.542108
  18.       two     0.282696 -0.087302
  19. baz   one    -1.575170  1.771208
  20.       two     0.816482  1.100230
  1. In [100]: stacked = df2.stack()
  3. In [101]: stacked
  4. Out[101]: 
  5. first  second   
  6. bar    one     A    0.029399
  7.                B   -0.542108
  8.        two     A    0.282696
  9.                B   -0.087302
  10. baz    one     A   -1.575170
  11.                B    1.771208
  12.        two     A    0.816482
  13.                B    1.100230
  14. dtype: float64
  1. In [102]: stacked.unstack()
  2. Out[102]: 
  3.                      A         B
  4. first second                    
  5. bar   one     0.029399 -0.542108
  6.       two     0.282696 -0.087302
  7. baz   one    -1.575170  1.771208
  8.       two     0.816482  1.100230
  10. In [103]: stacked.unstack(1)
  11. Out[103]: 
  12. second        one       two
  13. first                      
  14. bar   A  0.029399  0.282696
  15.       B -0.542108 -0.087302
  16. baz   A -1.575170  0.816482
  17.       B  1.771208  1.100230
  19. In [104]: stacked.unstack(0)
  20. Out[104]: 
  21. first          bar       baz
  22. second                      
  23. one    A  0.029399 -1.575170
  24.        B -0.542108  1.771208
  25. two    A  0.282696  0.816482
  26.        B -0.087302  1.100230

数据透视表

详情请参阅：数据透视表.

  1. In [105]: df = pd.DataFrame({'A' : ['one', 'one', 'two', 'three'] * 3,
  2.    .....:                    'B' : ['A', 'B', 'C'] * 4,
  3.    .....:                    'C' : ['foo', 'foo', 'foo', 'bar', 'bar', 'bar'] * 2,
  4.    .....:                    'D' : np.random.randn(12),
  5.    .....:                    'E' : np.random.randn(12)})
  6.    .....: 
  8. In [106]: df
  9. Out[106]: 
  10.         A  B    C         D         E
  11. 0     one  A  foo  1.418757 -0.179666
  12. 1     one  B  foo -1.879024  1.291836
  13. 2     two  C  foo  0.536826 -0.009614
  14. 3   three  A  bar  1.006160  0.392149
  15. 4     one  B  bar -0.029716  0.264599
  16. 5     one  C  bar -1.146178 -0.057409
  17. 6     two  A  foo  0.100900 -1.425638
  18. 7   three  B  foo -1.035018  1.024098
  19. 8     one  C  foo  0.314665 -0.106062
  20. 9     one  A  bar -0.773723  1.824375
  21. 10    two  B  bar -1.170653  0.595974
  22. 11  three  C  bar  0.648740  1.167115

可以从这个数据中轻松的生成数据透视表：

  1. In [107]: pd.pivot_table(df, values='D', index=['A', 'B'], columns=['C'])
  2. Out[107]: 
  3. C             bar       foo
  4. A     B                    
  5. one   A -0.773723  1.418757
  6.       B -0.029716 -1.879024
  7.       C -1.146178  0.314665
  8. three A  1.006160       NaN
  9.       B       NaN -1.035018
  10.       C  0.648740       NaN
  11. two   A       NaN  0.100900
  12.       B -1.170653       NaN
  13.       C       NaN  0.536826